Document 13548214

advertisement
Large scale synthesis and purification of iron responsive element RNA for crystallographic studies of
its interactions with iron regulatory proteins 1 & 2
by Brian John Eilers
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Biological Sciences
Montana State University
© Copyright by Brian John Eilers (2003)
Abstract:
Iron homeostasis is important to organisms in order to prevent various diseases, such as anemia and
hemochromatosis, a disease state caused by elevated iron levels that leads to increased levels of
reactive oxygen species and death due to stroke, heart attack, diabetes, cirrhosis of the liver or liver
cancer, usually by age 50. In addition, misregulation of iron has been associated with many diseases of
the central nervous system including Alzheimer’s and Parkinson’s. Thus, greater insight into the
moleuclar mechanisms of iron transport and homeostasis could lead to strategies to treat or prevent
these diseases. Iron regulatory proteins 1&2 are known to play a central role in the maintenance of iron
homeostasis. IRPs respond to cellular iron levels by binding to iron responsive elements present in the
mRNA corresponding to proteins involved in the transport, use and storage or iron. This interaction
serves to appropriately up or down regulate translation of these proteins. A major goal of the laboratory
is to obtain the crystallographic structures of the IRP-2/IRE complex. We have worked to express mg
amounts of rat IRP-2, and to synthesize mg amounts of the rat ferritin IRE. A mutant IRP-2,
IRP-2-ΔDD was created and initial expression trials indicated that it expressed at 0.5mg/L in a
Saccaromyces cerevisiae expression system. The IRE was produced by enzymatic synthesis using T-7
RNA polymerase. Two strategies were used. The first was a simple run-off transcription reaction.
Large scale synthesis of run-off IRE can yield up to 3 mg of RNA per 60 mL reaction. The second
method incorporated 5’ and 3’ flanking hammerhead ribozymes (HHR) in order to produce a
homogenous transcript. Initial small scale HHR cleavage experiments indicated that homogenous IRE
could be obtained from the HHR containing RNA. In both cases, the T7 transcription reactions were
purified by phenol/chloroform extraction, ethanol/isopropanol precipitation, and by size exclusion
chromatography. Size exclusion columns effectively removed the majority of the unincorporated NTPs,
while a Mono Q column was used to separate IRE from the HHR. Bandshift assays indicated that the
run-off IRE could bind to IRPs and was therefore active. LARGE SCALE SYNTHESIS AND PURIFICATION OF IRON RESPONSIVE
ELEMENT RNA FOR CRYSTALLOGRAPHIC STUDIES OF ITS INTERACTIONS
WITH IRON REGULATORY PROTEINS I & 2
By
Brian John Eilers
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
Biological Sciences
MONTANA STATE UNIVERSITY
Bozeman, Montana
July 2003
APPROVAL
of a thesis submitted by
Brian John Eilers
This thesis has been read by each member of the thesis committee and has been
found to be satisfactory regarding content, English usage, format, citations, bibliographic
style, and consistency, and is ready for submission to the College of Graduate Studies.
C. Martin Lawrence,
(Signature)
Date
Approved for the Department of Biological Sciences
7
Gwen Jacobs
(Signature)
- 11Date
Approved for the College of Graduate Studies
Bruce R. McLeod
(Signature)
Date
0 'S
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements for a master’s
degree at Montana State University, I agree that the library shall make it available to
borrowers under rules of the Library.
If I have indicated my intention to copyright this thesis by including a copyright
notice page, copying is allowable only for scholarly purposes, consistent with “fair use”
as prescribed in the U.S. Copyright Law. Requests for permission for extended quotation
from or reproduction of this thesis in whole or in parts may be granted only by the
copyright holder.
Signature
iv
TABLE OF CONTENTS
1.
INTRODUCTION...................................
I
Iron Homeostasis..................
I
Regulation in Cellular Iron Metabolism......................................................... ...I
Mechanism of Ferritin Translational Control.................................................... 3
Mechanism of Transferrin Receptor Translational Control............................... 5
Binding of IRP-I and IRP-2 to IREs................................................................. 8
Posttranslational Regulation of IRE-Binding Activity.....................................11
2.
CLONING AND PROTEIN EXPRESSION........................................................ 19
Expression of IRP-2 Degradation Domain Deletion Mutants...............................19
Introduction...................................................................................................... 19
Construction of Degradation Deletion Mutant................................................ 20
Subcloning IRP-2-ADD Constructs into Baculovirus Expression Vectors.....26
Gateway Cloning and Yeast Expression.......................................................... 29
Results.............................................................................................................. 32
3.
SYNTHESIS AND PURIFICATION O FIRE..................................................... 33
IRE Construction.................................................................................................. 33
RNase Control....................................................................................................... 40
T7 RNA Polymerase Expression and Purification................................................ 48
Introduction................................................... *................................................. 48
Expression of the T7 RNA Polymerase........................................................... 48
Purification of the T7 RNA Polymerase.......................................................... 49
T7 activity assay.............................................................................................. 49
Methods............................................................................................................ 51
Results..............
56
Discussion and conclusions............................
60
PCR Amplification of Template DNA................................................................. 61
Methods............................................................................................................ 61
Results..... .........................................................................................................67
In vitro T7 transcriptions.......... ............................................................................68
Results....... ..................
72
V
TABLE OF CONTENTS - CONTINUED
IRE Purification.................................................................................................... 75
Phenol/Chloroform Extractions....................................................................... 75
Nucleotide Removal......................................................................................... 77
Results.............................................................................................................. 84
IRE Quality Control.............................................................................................. 85
UV-absorbance................................................................................................ 85
Mini PAGE...................................................................................................... 86
Sequencing Gels............................................................................................... 86
Band Shift Assays.....................
89
Conclusions...................................................................................................... 90
4.
CONCLUSIONS............................ ...................................................................... 91
IRP-2 minus degradation domain......................................................................... 91
IRE........ ................................................................................................................91
Future Experiments...............................................................................................93
REFERENCES
94
vi
LIST OF TABLES
Table
Page
1.
Physiological effects of IKE-IRP interactions................................................. 2
2.
Nucleic Acid modifying enzymes................................................................... 43
3.
CPM and DPM counts for blanks and T7RNP preps, and initial rates of
T7RNP preps...................................................................................................56
4.
Relative specific activities of the T7RNP....................................................... 59
5.
PCR reaction using plasmid as template......................................................... 63
6.
General T7 transcription reaction conditions..............
7.
Current T7 transcription reaction conditions for run-off T7 transcript........... 70
8.
Applications of pH specific phenol mixtures.................................................. 77
69
vii
LIST OF FIGURES
Figure
Page
1. IRPs Binding to Ferritin mRNA............................................................................. 4
2. Interactions of IRP with Iron and IRE Containing mRNA..................................... 5
3. IRP Stabilizes Transferrin Receptor mRNA by Preventing its Degradation.......... 7
4. Consensus IRE Sequences...........................................
9
5. Regulation of IRP-1 and IRP-2..............................................................................13
6. Multiple Sequence Alignment of IRPs and Aconitase..........................................15
7. Diagram of Site Directed Mutagenesis by Overlap Extension............................. 21
8. Amplification of AB and CD Fragments by PC R ................................................ 23
9. PCR Amplified IRP-2 Fragments A D .................................................................. 24
10. IRP-2-ADD and IRP-2-ADD-Gly cut with Avr II and Bgl I I ............................... 25
11. Colony PCR of IRP-2-ADD and IRP-2-ADD-Gly................................................ 25
12. Restriction Digest of IRP-2-ADD and IRP-2-ADD-Gly for Insertion into the
pFastBacl Vector............................................................................................... ,..26
13. Colony PCR and Restriction Digest to Check for Insertion of IRP-2-ADD
genes into pFastBacl Vector................................................................................ 27
14. Procedure for the Bac-to Bac baculovirus expression system.............................. 28
15. Western blot of pelleted cells from primary viral stock to detect IRP-2-ADD....29
16. IRP-2-ADD expression trial.........................................................................
32
17. Wild type ferritin IRE sequence from ra t.... .........................................................33
18. Synthetic IRE appears to have multiple bands..................................................... 34
viii
LIST OF FIGURES - CONTINUED
19. The proposed HUR cleavage mechanism............................................................. 36
20. The HHR-IRE-HHR construct.............................................................................. 37
21. Mutant ferritin IR E ............................................................................................... 38
22. The pBPBE-3 vector and various primers that can be used to generate a variety
ofRNA.....................................................................................
39
23. Ambion7s RNase Alert Test kit procedure........................................................... 47
24. T7 transcription reaction scheme.......................................................................... 49
25. [a-35S]-UTP incorporation using the commercial or T7RNP-1.......... ;............... 57
26. [a-35S]-UTP incorporation using T7RNP-3......................................................... 57
27. [a-35S]-UTP incorporation using T7RNP-5......................................................... 57
28. Wild type IRE DNA fragment generated by BP-25-RIR and BP-34-RIR........... 64
29. Template DNA amplified with various primers................................................ ...66
30. Comparison of enzymatic synthesis versus chemical synthesis of IRE................ 73
31. Various T7 RNA polymerase transcription rxns of HHR-IRE-HHR................... 73
32. Sequencing gel run of HHR-IRE-HHR and run-off IRE transcription rxns......... 74
33. NAP profile of run-off IRE................................................................................... 78
34. Nap profile of HHR-IRE-HHR.................................
79
35. P4 BioGel column loaded with IOml of 30mer IRE............................................. 80
36. Mono Q run of HHR-IRE-HHR.....................................................................
82
37. Mono Q run of HHR-IRE-HHR with step gradient.............................................. 82
ix
LIST OF FIGURES - CONTINUED
38. Fractions (3, 15-25) corresponding to the peak containing NTPs from the
MonoQ run in Fig. 3 3 ...........................................................................................83
39. Fractions (26-34) corresponding to the peak containing IRE from the
MonoQ run in Fig. 3 3 ...........................................................................................83
40. Fractions (35-43) corresponding to the peak containing HURs from the
MonoQ run in Fig. 3 3 ...........................................................................................84
41. Cutting experiment on HHR-IRE-HHR constructs run out on a sequencing gel ..88
42. Bandshift Assay of synthetic IRE verses T7 run-off IRE being bound by IRP-I
and IRP-2.................. :.......................................................................................... 90
ABSTRACT
Iron homeostasis is important to organisms in order to prevent various diseases, such as
anemia and hemochromatosis, a disease state caused by elevated iron levels that leads to
increased levels of reactive oxygen species and death due to stroke, heart attack, diabetes,
cirrhosis of the liver or liver cancer, usually by age 50. In addition, misregulation of iron
has been associated with many diseases of the central nervous system including
Alzheimer’s and Parkinson’s. Thus, greater insight into the moleuclar mechanisms of
iron transport and homeostasis could lead to strategies to treat or prevent these diseases. 1
Iron regulatory proteins 1&2 are known to play a central role in the maintenance of iron
homeostasis. IRPs respond to cellular iron levels by binding to iron responsive elements
present in the mRNA corresponding to proteins involved in the transport, use and storage
or iron. This interaction serves to appropriately up or down regulate translation of these
proteins. A major goal of the laboratory is to obtain the crystallographic structures of the
IRP-2/IRE complex. We have worked to express mg amounts of rat IRP-2, and to
synthesize mg amounts of the rat ferritin IRE. A mutant IRP-2, IRP-2-ADD was created
and initial expression trials indicated that it expressed at 0.5mg/L in a Saccaromyces
cerevisiae expression system. The IRE was produced by enzymatic synthesis using T-7
RNA polymerase. Two strategies were used. The first was a simple run-off transcription
reaction. Large scale synthesis of run-off IRE can yield up to 3 mg of RNA per 60 mL
reaction. The second method incorporated 5’ and 3’ flanking hammerhead ribozymes
(HHR) in order to produce a homogenous transcript. Initial small scale HHR cleavage
experiments indicated that homogenous IRE could be obtained from the HHR containing
RNA. In both cases, the T l transcription reactions were purified by phenol/chloroform
extraction, ethanol/isopropanol precipitation, and by size exclusion chromatography. Size
exclusion columns effectively removed the majority of the unincorporated NTPs, while a
Mono Q column was used to separate IRE from the HHR. Bandshift assays indicated
that the run-off IRE could bind to IRPs and was therefore active.
I
CHAPTER I
Introduction
Iron Homeostasis
Regulation in Cellular Iron Metabolism
When iron is present as heme, iron-sulfur clusters, or in a direct relation with
proteins it has a key role in an extensive number of crucial cellular functions, such as
oxygen transport, mitochondrial energy metabolism, electron transport, deoxynucleotide
synthesis, or detoxification (I). The free iron (non-protein bound) is generally thought to
catalyze the formation of highly reactive free radicals, which are known to damage
membranes, proteins and DNA (I). Thus, it is extremely important for both the cells and
the organism to maintain iron homeostasis to ensure a healthy supply of iron, but at the
same time to prevent excess accumulation of iron. Anemia is usually induced by either a
nutritional deprivation or malabsorption in whole organisms that results from an
insufficient uptake of iron (I). The opposite occurs in a genetic disorder called
hemochromatosis, where iron overload occurs by pathologically increasing the amount of
iron uptake. Hemochromatosis will lead to permanent cell and tissue damage because the
extracellular iron binding capacity of transferrin as well as the intracellular iron storage
capacity of ferritin has exceeded safe limits (I). Disease states associated with hereditary
hemochromatosis include heart attack, stroke, diabetes, cirrhosis of the liver, and liver
cancer with death usually occurring by age 50. In addition, misregulation of iron is also
2
implicated in diseases of the central nervous system, including Parkinson’s, Alzheimer’s,
multiple system atrophy, and neuronal brain iron accumulation type I.
Intracellular iron levels control the expression of key proteins in the iron
metabolism of vertebrate cells at the transcriptional level. Specific mRNA-protein
interactions in the cytoplasm mediate this regulation. Some mRNAs form specific hairpin
structures known as iron responsive elements (IREs) (I). These IREs are recognized by
iron-regulatory proteins (IRPs), which have been formerly referred to as IRE-binding
protein (IREBP), iron regulatory factor (IRE), or ferritin repressor protein (FRP) (I).
Two IRPs have been identified, IRP-I and IRP-2, that control the rate of mRNA
translation or stability (I). Both IRP-I and IRP-2 demonstrate an ability to bind IREs
under conditions of iron deprivation, but when the iron supply to cells is increased they
become posttranslationaly inactivated (IRP-I) or degraded (IRP-2) (Table I) (I).
Single IRE in 5’ UTR
Translation inhibited
Translation not inhibited
Ferritin H-chain and L-chain
Ferroportin
Erythroid 5-ALA-synthase
Iron storage low
Iron storage?
Porphyrin synthesis
diminished?
Iron storage increased
Iron storage
Porphyrin synthesis normal
Mitochondrial aconitase
Mitochondrial succinate
dehydrogenase subunit b*
?
?
?
?
Multiple IREs in 3’ UTR
mRNA degradation
inhibited
Rapid mRNA degradation
Transferrin receptor
DMT-I
dCytb
0-amyloid precursor
Iron uptake increased
Iron transport?
Iron uptake low
Iron transport?
?
?
?
?
Table I: Physiological effects of IRE-IRP interactions.
♦Present in Drosophila melanogaster, but not evident in man.
3
The posttranscriptional (translational) control mechanisms resulting from
IRE/IRP interactions have provided information about cellular iron homeostasis and the
maintenance of free cellular iron at the junction between iron uptake, iron storage, and
iron incorporation into proteins. Since IREs are present in a variety of mRNAs that
function in either of these pathways by encoding the proper protein, the IRE/IRP
interactions affect every major aspect of iron metabolism.
Mechanism of Ferritin Translational Control
IREs were first discovered in the 5’ untranslated regions (UTR) of ferritin H- and
L-chain mRNAs (2). These IREs were found to mediate inhibition of ferritin mRNA
translation in iron-deprived cells, but when cellular iron is abundant ferritin synthesis is
not halted, this leads to an increase in the iron storage capacity of the cell (2). The
physiological significance is that this results in ai feedback regulation whereby a free iron
pool controls the formation of its own storage sites. This regulation is supported by the
fact that inhibition of ferritin mRNA translation (in vitro) depends directly on IRP-I
binding to the 5’ IREs in the ferritin H- and L-chain mRNA (3).
In the 1960s, iron regulation of ferritin expression was thought to be under
translational control. Activation of ferritin synthesis in iron-treated rats was shown to be
insensitive to transcriptional inhibitors and associated with a shift in ferritin mRNA from
translationally inactive messenger iibonulceoprotein particles to polyribosomes (4). This
suggested that a protein was binding to the 5’ UTR of ferritin mRNA and suppressed its
translation in iron depleted cells. During the 1980s, it was determined that the regulation
4
of ferritin translation was controlled by IRP-I binding to an IRE in the 5’ UTRs of the
ferritin H- and L-chain mRNAs (2). “The IRE is necessary for the posttranscriptional
regulation of ferritin expression by iron and suffices to confer IRP-mediated translational
control to reporter mRNAs in transfected cells” (I). Polyadenylylation of the mRNA
template is not required, at least in vitro, for IRP-mediated translational repression (5).
This indicates that the binding of IRP-I or IRP-2 to the IRE is sufficient to regulate
translation without requiring additional translation factors. Non-IRE sequences on either
side of the IRE may affect the structure and function of the IRE, and pre- and
posttranslational mechanisms have been shown, in vivo, to regulate ferritin expression
(6). The IRE/IRP complex inhibits mRNA translation by preventing the small ribosomal
subunit (the 43S translation preinitiation complex) from binding to the mRNA (7). The
43S preinitiation complex binds the 5’ UTR of the ferritin mRNA, and since the ferritin
IRE is located < 40 nucleotides from the 5’ end of the mRNA they are both competing
for the same site (6). Therefore, the IRE/IRP complex is acting like a steric inhibitor of
the 43S preinitiation complex binding (Fig. I and Fig. 2)
4os
:
43S Entry Site
Fig. I: Illustration showing how IRPs could prevent the ribosomal components
from binding to ferritin mRNA (I).
5
Bound IRP
Low Iron:
Translation
Repressed
5 UTR
elF4F
Coding Region
3 UTR
5’CAP
Free IRP
Newly
S ynthesized
Peptide
43S com plex
High Iron:
Translation
Activated
iOS^pibos^mal
Mammalian m RNAs that contain an IRE
their S'UTfi
W - - U , c ___,4,„
H- and L-Ferrltin
5-Am inolevulinate S y n th ase
Ferroportin
m -A conitase
Fig. 2: Interactions of IRP with Iron and IRE containing mRNA.
Mechanism of Transferrin
Receptor Translational Control
After discovering the IREs in the ferritin mRNAs, five similar motifs were
identified within the 3’ UTR of transferrin receptor (TfR) mRNA (8). The specific
location of these IREs agreed with two regions of about 200 bases each that are known to
give differential stability to the TfR mRNA depending upon the cellular iron levels (9).
Interaction between IRP-I and the TfR IREs was demonstrated in cells that were treated
with an iron chelator this caused stabilization of TfR mRNA and increased protein
expression (10). The opposite was seen when iron was added to the cells, the extra iron
inhibited IRE binding to IRP-I (within 2 hours) and degraded TfR mRNA (I). Through
analysis of numerous deletions and mutations in the regulatory regions it appeared that
6
ERP-I protected TfR mRNA once it was bound (I). Therefore when cells are iron
deficient more TfR is produced in order to absorb more iron through endocytosis of
transferrin. Just like iron storage, iron uptake is adjusted by a feedback control loop in
which a free iron pool controls its own size.
Differential expression of TfR was first reported in the early 1980s when
proliferating cells responded to variations in iron availability (11). Once the TfR cDNAs
were isolated, experiments in cultured cells showed an increase or a decrease in TfR
mRNA levels in the presence of iron chelators and iron salts respectively (12). Initially
TfR regulation was believed to be under transcriptional control, but early experiments
didn’t show this (13). It wasn’t until the 3’ UTR of the TfR mRNA were deleted that
translational control became evident (9). With the 3’ UTR deleted, TfR expression
increased and was unregulated in cultured cells. “Deletion mapping of the regulatory
sequences identified two necessary areas of about 200 nucleotides each, separated by
some 250 nucleotides” (I). These two regulatory sequences (not adjacent to the 3’ UTRs)
showed high sequence identity between human and rat mRNAs (94%) and between
human and chicken mRNAs (89%) (10). The conservation in sequence in these regions
was higher than that of the coding regions themselves. Since the regulatory regions of
TfR mRNA are more complex than that of ferritin mRNA, its structure can’t be predicted
with certainty, but the RNA stabilizing and destabilizing elements have been found.
Through the use of in vivo footprinting experiments the IREs are believed to adopt the
hairpin conformation, and in vitro binding experiments showed that at least four IREs are
available for ERP binding (14). When iron levels are low in the cell the binding activity of
7
IRE for IRP correlates with the stabilization of TfR mRNA. Through mutational studies
it was determined that if the IRE was mutated so that it could not bind IRP, the instability
of the TfR mRNA would increase (15). In addition, mutations or deletions in the
destabilizing regions would lead to a stable TfR mRNA (8,15). When the iron levels
increase in the cell the IRPs become inactive and the IRE/IRP complex dissociates, this
causes the destabilizing regions to be exposed (Fig. 3) (16). It is not known if the
destabilizing regions (once exposed) undergo a conformational change before the mRNA
is degraded. The actual mechanism of RNA degradation and the ribonucleases involved
remains unsolved for now.
Fig 3: IRP stabilizes transferrin receptor mRNA by preventing its degradation.
8 •
An extended regulatory network that operates through IRPs appears to connect
the synthesis of protoporphyrin IX (in erythroid cells) and mitochondrial iron sulfur
proteins to the availability of iron (I). There are several IREs that are located in the 5’
UTRs of the mRNAs that encode proteins in these pathways. These mRNAs function
similarly to those of the ferritin mRNA by acting as translationally controlled elements.
The IRE in 5-ALA synthase mRNA is a good example of a regulatory connection
between iron availability and heme synthesis for hemoglobin, which is a major iron
utilization pathway (17). Direct experimental evidence linking the inhibitory effect of
activated IRPs on the synthesis of 5-ALA-synthase is still lacking. However, with IREs
present in the mRNAs encoding for citric acid cycle enzymes there seems to be an
unexpected connection between iron and mitochondrial energy metabolism. How and
why this connection evolved remains largely unsolved, but the discovery of these new
IREs confirms an earlier evolutionary origin, tracing back to arthropods and mollusks
(18). The known network of IRE-regulated mRNAs may grow in the future as research
progresses.
Binding of IRP-I and IRP-2 to IREs
That the IRE is highly conserved for any given gene between species and is
similar between different genes containing IREs implies that precise structural constraints
in the binding of IRE to IRPs are required. The structure of the IRE consists of a stem
and loop with a five base paired stem of variable sequence followed by a six nucleotide
loop with a consensus 5’-CAGUGN-3’ (19). Beneath the paired stem there is an
unpaired cytosine that results in a small asymmetrical bulge (19). In some IREs this
9
bulge is drawn as a single unpaired C nucleotide, while in others the C nucleotide and
two other nucleotides oppose one free nucleotide. The C bulge could give a specific bend
to the IRE structure. Below the C bulge lies another base paired region that appears to
give stability to the IRE hairpin (Fig. 4) (I). Recently the NMR structure of a 30
nucleotide IRE with a single C bulge has been determined (20). The loop samples a wide
range of conformations, while the C bulge appears to be unstructured, and stems appear
well defined as expected.
G3
A2
G3
0«
Cl .
G5
A2
U4
Cl .
loop
G5
N*
N
N
N
N
N
•
•
.
•
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
C
C
G
C
U
N
N
N
N
5 '- N
.
♦
.
.
•
•
•
•
N
N
N
N
N - 3'
consensus IRE
with "ferritin"
bulge
N • N
N
N
N
N
N
5'-N
• N
• N
. N
• N
- N
- N - 3'
uppei stem
bulge
lower stem
consensus IRE
with single C
bulge
Fig. 4: Consensus IRE sequences, N represents any nucleic acid. The bulge
and the loop are conserved between species, while the stems differ between
Several IRE/IRP-1 binding experiments have been done to elucidate the structure
and sequence constraints of the IRE (I). Wild-type IREs have a high affinity of binding
(Kd ~ 10-30 pM) to IRP-1, but when nucleotides in the loop and bulge region are deleted
10
the binding affinity is severely impaired or destroyed (21). Point mutations disrupting the
upper stem were found to be nonfunctional; however compensatory mutations restoring
the base pairing were tolerated (21). Nucleotide exchanges in the first five nucleotides of
the loop, or within the bulge nucleotides of the IRE usually result in significant decrease
in binding to IRP-I (21). These types of mutations can dramatically affect iron
regulation in vivo, as in the case of dominant familial hyperferritnemia that leads to earlyonset cataract (22). In hyperferritnemia, the ferritin L-chain genes have an A2 to G2
mutation in the IRE loop (see Fig 4 for nomenclature and location); this lowers the
affinity for IRP causing excessive ferritin synthesis in the affected individuals.
In SELEX experiment, a pool of 16,384 IRE variants was selected in order to find
the best IRP-I binding sequences (23). The variants were mutated randomly in the six
nucleotide loop and C bulge. The consensus sequence selected from the top picks
corresponded precisely with the phylogenetically conserved IRE, but a number of single
and double mutations were tolerated and showed a 2 to 10 fold decrease in affinities (I).
These experiments provided strong evidence that the loop is likely to be structured
between nucleotides one and five (I). Some of the double mutations at the loop,
specifically C1 and G5 altered to U1A5 and to G1C5 (see Fig 4 for nomenclature and
location), showed affinities comparable to wild-type IRE (28). NMR analysis of the IRE
structure suggested that there are intraloop base pair interactions (20). While, a SELEX
approach and a direct screening method yielded IRE mutants and double mutants that
specifically bind to IRP-I and IRP-2 respectively (23). Differences in affinity between
11
ERP-I and ERP-2 for the various EREs allow subtle regulation of the various messages
involved in iron transport and homeostasis.
Posttranslational
Regulation of IRE-Binding Activity
IRP-I and IRP-2 were first discovered when they interacted with EREs in rat liver
cytoplasmic extracts, they have since been found in most cell lines and tissues (24). Both
IRP-I and ERP-2 have strong affinities for the wild-type IREs. They are active when iron
levels are low in the body, but lose their affinity for the ERE when iron levels are high
(25). Through the uses of western blots and gel retardation assays it was determined that
ERP-I is ubiquitous in all vertebrate tissues (25). IRP-I expression varies between cells
lines from 50,000 (mouse) and 100,000 (human) molecules per cell, with only 10%
actively binding ERE, but with the addition of an iron chelator BRP-I activity can reach
100% (I). This would indicate an all-or none event, but this is not the case. IRP
regulation depends on the equilibrium between IRE/IRP complex and free IRP. This
allows the cell to have the flexibility to adapt to changes in the supply of iron, and to
respond to iron requirements (e.g. cell proliferation).
The total amount of BRP-I in the cell doesn’t have to change in order to alter the
activity of IRP-1. Through various experiments it was determined that all of the IRP-I
whether it was active or not, could be converted into the active ERE binding form by
adding 2% 2-mercaptoethanol (26). This meant that an estimate of the total amount of
ERP-I present could be obtained, and this provided the first solid evidence that changes in
IRP-I activity resulted from an iron-induced posttranslational modification. In order to
12
explore the iron-induced posttranslational modification, IRP-I was purified. IRP-I
consists of 889 amino acids with a predicted molecular weight of 98 kDa, exists as a
monomer in the cytoplasm, and is homologous to mitochondrial and bacterial aconitases
(14). Aconitases are iron sulfur proteins that interconvert citrate to isocitrate, so it was
predicted that IRP-I might display some aconitase activity as well (27). By adding
ferrous iron and sulfide to iron depleted cells, IRP-I converted into a cytoplasmic
aconitase by binding a 4Fe-4S cluster (27). The 4Fe-4S cluster binds to IRP-I at three
cysteine residues; the residues are located at positions homologous to other aconitases
(27). When the 4Fe-4S cluster is present, ERP-I cannot actively bind IRE and dissociates
from mRNA. However, when the 4Fe-4S cluster is not present the apoprotein (iron
deficient form) can actively bind ERE (27). The ability of IRP-I to convert between two
different forms illustrates how the cellular iron sensing mechanism can control the
interactions between IRP-I and IREs. By mutating the three cysteines necessary to bind
the 4Fe-4S cluster to serine, IRP-I reverts back to the apo-form and loses its aconitase
activity (28). Since the structure of IRP-I is unknown, the crystal structure of
mitochondrial aconitase can give clues to its possible the structure. Presumably, IRP-I is
composed of four domains with a deep cleft forming between domains 1-3 and domain 4,
with domain 4 connected to domains 1-3 by a flexible linker (Fig. 5) (I). From the
aconitase structure and various UV crosslinking studies, the ERE binding site appears to
involve several domains that cover the length of the IRE stem loop (29). The actual
mechanism regulating the interconversion of IRP-I and cytosolic aconitase by cellular
iron remains largely unsolved, as do the details of the interaction between IRPs and IREs.
13
Also, the IRP-I gene knockouts in mice show no discern able phenotype (Bioiron 2003),
but mice with IRP-2 gene knockouts show various neural disorders (30). Ultimately, the
crystal structure of the IRP-1/IRE complex will help to answer these questions.
Active IRP
IRP-I
Inactive IRP
Active
Aconitase
IRP-2
Fig. 5: Regulation of IRP-I and IRP-2. In iron replete cells, IRP assembles a cubane
Fe-S cluster that is liganded via Cys 437, 503 and 506. In iron replete cells, IRP-2 is
targeted for degradation via the degradation domain whereas IRP-I is stable and active
as a cytoplasmic aconitase.
Even less is known about IRP-2 than IRP-1. IRP-2 is expressed in a variety of
tissues but it is far less abundant than IRP-1, from gel retardation assays IRP-2 is strongly
expressed in the intestine and brain (30). When looking at the relative transcription
levels of the IRPs it was determined that IRP-2 had higher levels in the brain and IRP-I
had higher levels in the heart and liver (30). Once the full length cDNA of IRP-2 was
cloned it gave great insight into its function. Human IRP-2 is 57% identical and 79%
14
similar in amino acid sequence to human ERP-I (25). Purified IRP-2 is slightly larger
than ERP-I with a predicted molecular weight of 105 kDa, due to the presence of a 73
amino acid sequence (31). The 73 amino acid sequence or degradation domain is not
located in IRP-1, but highly conserved in IRP-2 between species (25). IRP-2 contains the
necessary cysteines residues to bind a 4Fe-4S cluster, but it does not have aconitase
activity (32). It remains unclear if IRP-2 can accommodate a 4Fe-4S cluster. Unlike
IRP-1, IRP-2 is proteolytically degraded in iron depleted cells when iron is added, due to
the unique 73 amino acid sequence. When the 73 amino acid sequence is deleted from
IRP-2, the protein does not go undergo proteolytic degradation (31). However, when
IRP-I has the 73 amino acid sequence inserted it goes through the iron dependent
degradation, implying that the degradation domain is both necessary and sufficient to
catalyze iron mediated degradation (31). A multiple-sequence alignment was performed
on human IRP-1, human ERP-2, and pig mitochondrial aconitase (Fig 6). The alignment
also indicates the domains, and critical residues that are conserved.
IRP-2 degradation is controlled by cellular iron levels and degradation appears to
occur in the proteosome because it is prevented when specific proteosome inhibitors are
used. IRP-2 becomes oxidatively modified and ubiquitinated in vivo and is degraded by
the proteosome (33). It was determined that when IRP-2 is exposed to iron, oxygen, and
DTT it was the degradation domain that became oxidatively modified (33). Using mass
spectroscopy it was determined that the cysteine residues in the degradation domain were
being converted to a product 2 atomic mass units less than the cysteine (33).
IIu .
IR P -I
Hu. IR P-2
P ig A cn .
3 3 ---------- N P F A lrL A E H D PV Q P QKKF FN LNKL E DS RY GR L P F 3 IR V L LEAAIRHC DE T LVKK QDIENILIINNVTQIIKN
1 EV PFK PARVILQD FTGVP AVVD F AAHRD AVKK LG GD PE K IN
3DA PKAG YAFEY LI E TL N - DS SHKKF FD V SK L -G TK Y D V LPY SIRVL L EA A VRHC D G F LMKKED V H N ILDWKTKQ S D------- V E V P F F PARVILQD FTG I P AMVD F AAHRE AVKTI G GD PE K VH
RAKVAHS KFE PHEYIRYD L L E K N ID IV R K R IN R P LTL S E K IVYCHLDD PANQEIERCKTYLRLRPD RViHQDATAQHAHL Q F IS S
CL PK
Domain I
H u.
IR P -I
P ig A cn
P VC PADL V ID H S IQ VD FN R -----------------------------------------------------------------------------------------------------------------------------------------------------------------RADSLQK nQD LZFERNREPFEFIK N GSQ AFH
PAC PIDLTVDHS L Q ID F S K CAIQNAP NP GG GD L QKAGKl SP LKVQPKKLPCRGQTTCRGS CD SGEL GRNSGTFSS Q IE N T P IL C P FH LQ P V IE PETV LK D QEVIFG RNREPLQFrK NS SRVLK
V A V P ST IH C D H IIE A Q L GG----------------------------------------------------------------------------------------------------------------------------------------------------------------------------E KD LRRAKDINQ E VYNF LATA CAKYG
H u. IR P -I
H u. IR P -2
P ig Ac n .
N H R I I P P G SG lIH O V D L E Y LARVVFDQDGYYYPDSLVGTDSHiTM ID G L G IL G V G V G G IEA l AVHLGOP IS M V L P Ov IG Y R LHGKPHPLV TST D IV L T IT K H L S O VGW GKFVEFFGPGVAOL
S V A V IP PG 7 g | a h Q ID LEY LSRV V F E E K D L IF P D S W G T D S H jTMVNGLGILGVGV G G IE T IAVH LG LPV SLTLPEW G C E L T G 3 SN PFV TSID VY LG ITK H LRQ VGVAGKFVEPFGSGVSQL
- V G n rR P G S G iIH O I LLENYAYPGVL-------------------L I GTDSH#PHGGGLGGICIGVGGADAYDVHAGIPNELKCPKTIGVKLTGSLSGNTSIKDVILKVAGILTVKGGTGAIVEYHGPGVDS I
H u. IR P -I
H u. IR P -2
P ig A c n .
3IADRATIAHMC PEYGATAAFFP VDEVSITY LV Q T GRDIEKLKYIKKYLQA VGHFRD FNDPSQD PD FTQW EIDLKTVVPCCSGPKRPQDKVAVSI'HKKDFESLLGAKNTEQGFEG FQYAPEH
3 1VDRTTIAHMC PE YGAI L S F FP VDN VT LKEL E H I GF SE AK L E S HETY LKAVKL FRND QNSSGEPEYSQVIQINLNSIVPSVSGPKRPQDRVAVTDM KSDFQALLNEK-----V GFEG FQ LAAEK
SCTGHATIOniC-AE IG A T T SV FPYNHRHKKYL SKT GRAD----------------1 ANLADEFKD ELVPDPGCHYD Q V IE IN L S E L K P H IN G P F T P - D - I A - — H PViEVGSVAEK----------------------------------
H u. IR P -I
H u. IR P -2
P ig A cn .
HNDHKT F IY - D - - F T lA H G S V V IA A ITSCTNTSNPSVH1GAGLL AKKAVDA GL HVHP T I KTS L S PG SG VVTYY LQESGVM’ YLSQLGFDWC-Y GCHTCIGNSGPLPEPVVEAITQGDLYAVG7
Q K D IV SIH Y E GSEYK1SHGSVVIAAVISCTIINCNP SVHIAAGLL AKKAVEAGLRVKP T IR T S LS PGSGHVTHYLS SS GVL PYLSK LG FEIVC-YGCSTC VGNTAPLSDAVLNA VKQGDL YTC GI
------------------------------ E GNPLD IRV G LIG SCTN SSY ED -H G R S AAVAKQA LA H G L EC K --SQ FT I T PG S E Q IR A T IE RI<GYAQVLRDVGGI V L iN A C G P C IG ---------------QNDRKDIHC GE KNT1 7
H u. IR P -I
HU. IK P -2
P ig A c n .
L S G N fc F E W V E PN T R A N -TL ASP PL V IA Y A l AGT I R ID FE KEP LGVN AKGQQ VF LKDINPTRD EIQ A VER QYVIP GH F K IV Y Q X IE TVNE SNNAL A -T P SDK LFFHN SK STYI ES PP F F E N l
L S G N fc FE UK LLDU V R A N -IL A S P PL W AY AiAUTVN ID M JlE P LGTD P l GKti IY LHDIWPSKE EV HK VEE EfcV l LSfl F K iL KDXlEHGNKklilN S lE -A P D S VLFPIlIDLKSTYiKC P S IF D K j
T S Y l t y F IW N IA N P E T H A FV T SPEIV TA L A IAG T LK FHPETDFLTG-KDGKK FK L E i - - P D AD EL PRAEFD ---------- P GQDTYQSPPKD S-SG Q R V D V SP---------- T S Q R L Q L L E P --F D K N D
Hu. IR P-2
Degradation Domain
Domain 2
Domain 3
Linker
H u. IR P -I
IIu . IR P -2
P ig A cn .
TLD L 0 P P K S IV IA Y V 1L N L GD SVTTD H I SP AGNIARNSPAARYL TNRGLTPRE FNSYGSRRGNDAVKARGTFANIRILNRFL NKQAP Q T IH IPS G E IL D Y FD A iERY Q OA G LPLIV LA GKE YG
TKE P IA L Q A IEKAIIVI L YL GD 3 VTTD H I 3 P AG 3 IA FN 3 AAAK YL TNRGLT PREFNSYGARRGND AVflTR GT F i N I KL FNK 7 1 GK P AP K T IIIP P 3 GQTL DY FE A iE LYQKE G I P L I I LA GKKYG
GKD------------ L E E L Q IIIK V K G K C TTD H I SAAG------------------- P NL KFRGHLD N I SNNL LI G A IN -IE NR-KA N S-V RD AV T QE FG P V P ----------------------------- D T iR YYKQ HG I RN W I GDE N YG
Domain 4
H u. IR P -I
H u . IR P 2
P ig A cn.
AGflSRDNAAKGI FLLG IK AV LA ESY ERIH RSN LV G HG V rPLEY LPGENA D ALGITG Q ERY TIIIPEN LK PQ M KV Q VK LD T
GXTF QAVMFFD T I'VE L7YFLNG G IL N Y H IRKHAK 8 8 9
5 C y i ’ DNAAKCrYL LCV K A 7LA ESY EK IH K D H LIC IC I a DLQFL rC E N A D S L C L S C M T F S L T F rE E L S rC IT L N IQ T S T
CXVF SVI A S FE DDVE I ? LYKHC CL LN FVADK FS
963
I G flS R E H SA L E IR H L G G R A IIT K SF A PIH E IN LK EO G L lPLTFA D PA D Y NK IH PV — D K LTIO G L K D FA PG K IL K C IIK H PNGTQETIL LNET FNETQ IE NFRiGSALNRMKELQQK 7 5 4
Fig. 6: A multiple-sequence alignment of Human IRP-1, Human IRP-2, and Pig Aconitase is shown. Aconitase active-site
residues defined in crystallographic studies are indicated in red, additional active-site residues conserved in all members of
the Fe-S isomerase family are indicated in blue, residues highlighted in grey indicate aconitase active-site residues that are
not conserved in IRP-2, residues highlighted in yellow indicated residues critical for binding the 4Fe-4S cluster.
16
The loss of the cysteine residue required only micromolar concentrations of iron,
with 50 micromolar iron reducing the number of cysteines by one (It is interesting to
note, however, that iron concentrations within the labile iron pool (“free iron”) are
believed to be on the order of I micromolar, 50 fold less than that of the iron
concentrations used in this study (33). Thus the physiological significance of this study
is uncertain.) In addition to the cleavage of the cysteine residues, it was determined that
two lysine residues were also susceptible to hydrolysis, which indicates that they are also
near the iron-binding site. By mutating the first cysteine to an alanine, the rate of
oxidative modification decreased by over 5.0% and by mutating the second cysteine the
rate is decreased to greater than 80% (33). The initial step in the oxidative modification
was hypothesized to convert cysteine to dehydrocysteine, which is 2 atomic mass units
less than cysteine; subsequently the dehydrocysteine is converted into aminomalonic acid
(33). The conversion of the cysteine residues to acidic aminomalonate could induce
conformational changes in the degradation domain that might be important in the
recognition of IRP-2 for ubiquitination and degradation by the proteosome (33). The first
step in marking IRP-2 for proteolytic degradation is hypothesized to be binding of ferric
iron that is reduced by a cysteine to the ferrous state (33). This then allows molecular
oxygen to bind. Abstraction of a second hydrogen would lead to dehydrocysteine and a
ferric hydroperoxy or ferryl moiety that could further oxidize the dehydrocysteine (33).
The ferric ion is then released to bind to another IRP-2 molecule and catalyze another
oxidation cycle.
17
It appears that heme also plays some part in the degradation of IRP 2 by binding
to the 73 amino acid sequence and causing instability (I). Experiments indicated that
heme, rather than free iron, could also catalyze degradation of IRP-2 (34). The
experiments determined that while free iron can degrade 80% of the IRP-2, the presence
of inhibitors showed that the heme was not just a source for free iron (34). This
suggested that the iron must first be incorporated into the heme or heme like compound s
before it can degrade the IRP-2 (34). Recently, experiments showed that when IRP-2 was
oxidized in the presence of heme, the IRP-2 was targeted for ubiquitination by the RING
finger protein HOIL-I (34). Ubiquitination is carried out by a series of reactions that is
catalyzed by the E l, E2, and E3 family of enzymes, more specifically; the E3 RING
finger ligases target specific recognition signals (34). It is clear from experimental data
that the degradation domain has evolved into an efficient iron sensor. It should also be
noted that mouse IRP-2 knockouts overexpress ferritin and develop progressive
neurodegenerative diseases (35). These mice develop progressive tremors, proximal
muscle weakness and ataxia. Experiments indicate that the neurodegeneration in the
mice is caused by excess iron in certain areas of the brain due to the misregulation of
ferritin synthesis (35). Iron accumulation has been implicated in the progression of many
human neurodegenerative diseases such as Parkinson’s disease, Alzheimer’s disease,
multiple system atrophy, and neuronal brain iron accumulation type. I (35). Therefore, it
is important to determine the mechanisms of IRP-I and IRP-2 regulation as they might
aid in the design of potential therapeutics for these neural diseases. Obtaining the crystal
18
structure of ERP-2/IRE complex will provide structural insight into its role in iron
metabolism, and for the design of specific drugs.
19
CHAPTER 2
Cloning and Protein Expression
Expression of IRP-2 Degradation Domain Deletion Mutants
Introduction
IRP-2 contains a 73 amino acid sequence or degradation domain that is not
present in IRP-1. The degradation domain, as its name implies, targets IRP-2 for
degradation when iron levels in the cell are high. While the exact mechanism for how the
degradation domain signals degradation is not presently known, there are several
hypothesizes; they include iron catalyzed oxidation of cysteine that are then recognized
by a specific ubiquiton ligase complex; binding of heme by the degradation domain and
subsequent recognition by a ubiquiton ligase; or other signals for degradation that could
involve phosphorylation. Removal of the degradation domain from IRP-2 could
conceivably stabilize IRP-2, leading to increased yield of over expressed IRP-2. In
addition, the degradation domain deletion mutant might be more amenable to
crystallization since the degradation domain may contain intrinsically disordered
residues. A major goal of the lab is to obtain crystallographic structures of IRP-2; if the
degradation domain were disordered this might interfere with the crystallization process.
Therefore, we decided to create two mutants of IRP-2, one where the domain is deleted
completely, and another in which the domain is replaced with two glycine residues. The
rationale for adding of two glycine residues is to compensate for any flexibility that was
lost when the degradation domain was deleted.
20
From the literature on IRPs, it was determined that DRP expression in bacteria
yielded mostly non-soluble protein in the form of inclusion bodies, while baculovirus and
yeast expression systems yielded some soluble IRP-2 protein. In order to obtain
crystallographic structures of protein, the protein must be soluble. Therefore, the
laboratory decided to use a baculovirus expression system from Gibco BRL, and a yeast
expression system from Invitrogen in order to obtain soluble DRP-2 minus the degradation
domain.
Construction of Degradation Deletion Mutant
Construction of these two mutants was performed with site directed mutagenesis
by overlap extension (36, 37). Site directed mutagenesis by overlap extension involves
the generation of DNA fragments in one round of PCR that can be fused together in
another round of PCR (36). Two sets of primers are used, primers A and D allow the
whole gene to be amplified (outside primers), while primers C and B have
complementary sequences to each other, they contain the desired changes, and anneal to
sites within the gene (36). In the first round of PCR, primer A anneals to the 5’ end of
the gene, while primer B anneals to sequence right before the degradation domain (36).
In another PCR reaction primer D anneals to the 3’ end of the gene while primer C
anneals to the sequence right in behind the degradation domain (36). These PCRs will
generate two fragments AB and CD; fragment AB contains the sequence right before the
degradation domain, while CD contains the sequence right after the degradation domain
(36). In the final round of PCR, the AB fragment and the CD fragment will anneal to
each other because the B and C primers contain complementary sequences, once the two
21
fragments are annealed the sequence can be extended and amplified using primers A and
D (Fig. 7) (36).
t
CO
(3)
I
I
MUTANT FUSION PRODUCT
Fig. 7: Schematic diagram of site directed mutagenesis by overlap extension. The DNA
and primers are represented by lines with arrows indicating the 5’ to 3’ direction. The
black box represents a site that can have a mutated, deleted, or inserted product.
Elizabeth Leibold (University of Utah) graciously provided the lab with the rat
IRP-2 gene (in the pHT B2 YES vector). Since, the pHT B2 YES vector is a yeast
expression vector the IRP-2 gene was recloned into an E. Coli expression vector pUC19.
The IRP-2 gene was now in two vectors pHT B2 YES and pUC19.
22
George Gauss (Montana State University) helped to setup the experimental
procedure needed to perform site-directed mutagenesis. To create the degradation
domain deletion mutant (without extra glycine residues) the AB and CD fragments must
be generated. The PCR reaction mixture used to amplify the AB fragment was as
follows; I ng of template DNA (pHT B2 YES, IRP-2 gene from rat), 0.5 pM primer A:
GHG 120(5’ CCGACAGGATCCACCATGGACTCCCCAAGTG C 3’), 0.5 pM primer
B: BJE 102 (5’ aacaccgtctcaggttcTTTACTGAAGTCAATCT 3’), 0.2|iM dNTPs, 0.03
U/|il Taq polymerase, 0.00006 U/pl Vent polymerase, 50mM Tris pH 9.2, 16mM
(NH4)2SO4, 1.75 mM MgCl2, 0.1% Triton X-100. The reaction mixture was placed in a
thermocycler with the following parameters; [initial 94°C 120 seconds], 94°C 20
seconds, 55°C 30 seconds, 72°C 60 seconds repeated 35 times, 12°C 5min, and 4°C hold.
The PCR reaction mixture used to amplify the CD fragment was as follows; I ng of
template DNA (pHT B2 YES, IRP-2 gene from rat), 0.5 pM primer C: BJE 101 (5’
gaacctgagacggtgttAAAAAATC 3’), 0.5 pM primer D: BJE 201 (5’ AATTTCACAGCTT
TAAGGTA 3’), 0.2|iM dNTPs, 0.03 U/ul Taq polymerase, 0.00006 U/ul Vent
polymerase, 50mM Tris pH 9.2 ,16mM (NH4)2SO4, 1.75 mM MgCl2, 0.1% Triton X-100.
The reaction mixture was placed in a thermocycler with the following parameters; [initial
94°C 120 seconds], 94°C 20 seconds, 55°C 30 seconds, I l 0C 60 seconds repeated 35
times, I T C 5min, and 4°C hold. PCR reactions were run out on a 1% agarose gel to
check for correct fragment size (Fig. 8).
23
Fig. 8: Amplification of AB and CD fragments by PCR. PCR reactions were run out on
a I % agarose gel for 45 min. at 100V. The gel was ethidium stained and analyzed under
The same reaction conditions that were used to create the AB and CD fragments
are also used to create the AD fragment. The only exceptions are: primer A: GHG 120
(0.5 pM), primer D: BJE 201(0.5 |iM), IOOng of fragment AB (gel purified), and IOOng
of fragment CD (gel purified). The two fragments anneal together and act as the template
DNA because the ends of the BJE 101 and BJE 102 primers base pair together (small
case lettering). The thermocycling was under the same conditions as the AB and CD
fragments except that 72°C extension time was increased from 60 seconds to 120
seconds. The PCR reactions were run out on a 1% agarose gel to check for correct
fragment size (Fig. 9).
The same procedure that was used to create the degradation domain deletion
mutant (without glycine residues) is used to make the degradation domain with the two
glycine residues, except primer B would be BJE 104 (5’ gtctcaggttctcctccTTTACTGAA
GTCAATCT 3’) and primer C would be BJE 103 (5’ ggaggagaacctgagacGGTGTTAAA
AAATCAAG 3’). The result of these PCR experiments lead to obtaining amplified IRP2 gene with deletion of the degradation domain with or without glycine residues added.
24
IO O O b D
I OOObp
Fig. 9: PCR amplified IRP-2 fragments AD with and without extra glycine residues. PCR
reactions were run out on a 1% agarose gel for 45 min. at 100V. The gel was ethidium
stained and analyzed under a UV light.
Once the degradation domain was deleted from the IRP-2 gene, it was then
possible to subclone the gene using traditional methods. The primer on the far 3’ end was
engineered to include a natural enzyme restriction site (Avr II) located = 100 bases
behind the degradation domain, and the primer on the 5’ end was already close to a
natural restriction site (Bgl II) located = 370 bases from 5’ end, this meant that the newly
cloned portion of IRP-2 could be ligated into the original IRP-2 gene. This was done by
using Avr II and Bgl II to cut pUC19 (with IRP-2 gene), and the AD fragments (created
by PCR). The digested products were gel purified and then ligated together to create the
IRP-2 gene lacking the degradation domain (IRP-2-ADD) and the IRP-2 gene lacking the
degradation domain plus two glycines (IRP-2-ADD-Gly).
25
I Kb
3000bpt
ladder
Vector cut with
AVR 11 and Bgl 11
pUC19
with
^IRP-2 gene
1000b
Fragments AD cut
with AYR U and Bgl II
Fig. 10: IRP-2-ADD and IRP-2-ADD-Gly (fragments AD) cut with Avr II and Bgl II.
DNA ladder starts at 5kb.
Fig. 11: PCR of E. coli colonies to check IRP-2-ADD (BJE 301), and IRP-2-ADD-Gly
(BJE 302) for the degradation domain deletion.
E. coli (XL-1 blue strain) were transformed with the ligation products, and plated
out on LB selection media with antibiotic resistance (ampicillian 100/xg/^l). Plates were
26
incubated overnight at 37°C, and colonies were picked to use for DNA minipreps. The
minipreped DNA was then used for sequencing reactions. Sequencing Reactions were
sent out to Iowa State University Sequencing Facility for analysis. Only the newly
cloned portion of gene needed to be re-sequenced (~ 800 bp) versus re-sequencing the
entire gene (~ 2800 bp). Two clones without the degradation domain were now
generated; IRP-2-ADD (BJE 301), and IRP-2-ADD-Gly (BJE 302).
Subcloning IRP-2-ADD constructs
into Baculovirus expression vectors
The subcloning procedure was devised by George Gauss (Montana State
University). The vectors IRP-2-ADD (BJE 301), and IRP-2-ADD-Gly (BJE 302) along
with the pFastBacl vector (Gibco BRL) were digested with Bam HI and Xba I. The
appropriate bands were cut out of an aragose gel and ligated together. This generated
mutant IRP-2 genes that were now in the pFastBacl vector IRP-2-ADD (BJE 303), and
IRP-2-ADD-Gly (BJE 304).
Fig. 12: Restriction digest of IRP-2-ADD (BJE 301), and IRP-2-ADD-Gly (BJE 302) for
insertion into the pFastBacl vector. Top band of BJE 301 and BJE 302 is the IRP-2
gene with the degradation domain deleted, while the bottom band is vector pUC19.
27
Restriction Digest
of BJE 303 and BJE 304
Fig. 13: Colony PCR and Restriction Digest to check for insertion of IRP-2 degradation
domain deletion genes into pFastBacl vector.
Using standard protocols found in the BAC-to-BAC baculovirus expression
systems manual from Gibco BRL, the pFastBac I vector was transformed into competent
DHlOBac E. coli cells that contain the baculovirus genome (Bacmid), resulting in
transposition of the IRP-2 gene from pFastBacl into the Bacmid. The cells are then
plated out under antibiotic resistance. Colonies containing the IRP-2-ADD inserts will be
white, while colonies that do not undergo transposition retain the P-galactosidase gene
and appear blue. One white colony was picked and underwent a mini-prep for high
molecular weight DNA. The DNA was then used to transfect Sf9 insect cells and virus
was grown for 72 hours (Fig. 14). The primary viral stock was centrifuged to pellet the
cells, and the cells under went a Western blot to test for IRP-2 expression (Fig. 15).
PFastBhc donor pbsm id
TransIorinatDn ,
IQ
A
0-:;1
J
Com petent DH10B*: E ooA Cells
Transppsilion ^
Anlbldlc (Seleoicn "
I
I
I
r
I
jsj)
©
X
pPor
E. cotHLacT)
Containing Recombinant Bacmid
I
Recombinant Gene Expression
o r Viral Amplification
Fig. 14: The procedure for the BAC-to BAC baculovirus expression system. The pFast Bac-I plasmid with gene of
interest gets transposed into the baculovirus genome (Bacmid). Purified bacmid is used to infect insect cells and produce
virus. The viral stock is then used to infect insect cells and exoress orotein
29
The initial results from the expression study showed that the mutant IRPs expressed at the
same level as wild-type IRP-2 in baculovirus (O.Smg/L, George Gauss). From this result,
it was decided that IRP-2 expression in baculovirus should be put on hold.
Wild Type
IRP-2
IRP-2-ADD
pHT B2 Yes
with
IRP-2
Fig. 15: Western blot of pelleted cells from primary viral stock. cc-His antibody used
Gateway Cloning and Yeast Expression
The Lawrence lab has now made the switch from traditional cloning methods to
the Gateway Cloning System by Invitrogen. Therefore, George Gauss (Lawrence Lab,
Montana State University) cloned the IRP-2-ADD (BJE 301) and IRP-2-ADD-Gly (BJE
302) vectors into the Gateway Entry Vector.
The entry vector allows the user to transfer the desired gene of interest into any
expression vector in the Gateway system {Invitrogen). In the case of IRP-2 degradation
mutants, a yeast expression vector (pDest52) was used because native IRP-I and IRP-2
were expressing at sufficient levels (l-3mg/L), and yeast culture is significantly easier
than protein expression using the baculovirus system. IRP-2-ADD and IRP-2-ADD-Gly
30
were moved into pDest 52 using the Gateway site specific recombination system. A
yeast cell line was transformed with each of the pDest52-IRPADD vectors following the
EasyComp Yeast kit by Invitrogen. The procedure involves mixing the DNA, yeast
competent cells, and Easy Comp Solution IH {Invitrogen) in a microfuge tube and
incubating it at 37°C for one hour. The cells were then plated On -Ura plates containing
dextrose and incubated at 30°C for 2-3 days. A colony was picked and grown in 50 ml
starter culture containing dextrose. Then 25 ml of the overnight culture was added to
expression media expression containing galactose rather than dextrose. The yeast cultures
then underwent expression for 24 hours. From previous experiments on IRP-I and IRP-2
in yeast it was determined that 24 hours was sufficient in order to see expression of
protein. The yeast cultures were centrifuged to pellet the cells, and cell pellets were
frozen at -20° C. Some additional things to consider for yeast expression are:
1. All work should be done in a sterile hood to avoid contamination of the yeast
2. Rafinose can also be used instead of dextrose for overnight cultures
3. Galactose, glucose, and dextrose can be autoclaved
4. Rafinose must be filter sterilized
5. Overnight starter cultures may be spun down to remove sugars before adding the
cells to the expression media
6. After 48 hours yeast cells become difficult to break open.
7. Yeast cells break open best when using a bead beater
31
The yeast cell pellets were then thawed on ice, and resuspended in lysis buffer (50
mM Tris-HCl pH 8.0, 300 mM NaCl, 5 mM imidazole, 2.5 mM 2-ME, 0.1 mM EDTA
pH 8.0, and 0.1 mM PMSF). The resuspended cells, 60 grams of dry glass beads, and
additional lysis buffer were added to the bead beater container. The cells were broken
open by grinding for 30 seconds and allowed to cool for 60 seconds, this was repeated 5
times. The lysate was then separated from the beads, and centrifuged at 4°C for 20
minutes at 17k fpm. The lysate was then loaded onto a gravity flow Ni-NTA column
(3ml of resin). The column was then washed 3 times with 3 ml of lysis buffer, and then
washed 2 times with 3 ml lysis buffer plus 20 mM imidazole. The protein was eluted
from the column in 3 ml of lysis buffer plus 250 mM imidazole. The fractions were then
run out on a 10% SDS-PAGE gel for 45 min at 225 V.
Anayalsis of the gels indicated that the IRP-2-ADD construct had expressed
protein at the correct molecular weight (97 kDa) that eluted off the Ni-NTA column
(Fig. 16). The IRP-2-ADD-Gly construct showed protein expression in the cell pellet, but
no protein was seen in the elution fractions. According to Bradford assays, IRP-2-ADD
expressed at 0.5mg/L. George Gauss and Philippe Benas were able to express wild-type
IRP-2 at Img/L and IRP-I at 2-3mg/L respectively in yeast cells (saccaromyces
ceriviciae).
32
I
2
3
4
5
6
7
8
9
IRP-2-ADD
10
100— >
Fig. 16: IRP-2-ADD expression trial. Lane I: protein standards. Lane 2: cell pellet.
Lane 3 Ni-NTA column flowthrough. Lanes 4-7 column washes. Lane 8: first elution,
Lane 9: second elution, Lane 10: third elution. Fractions were then run out on a 10%
SDS-PAGE gel for 45 min at 225 V.
Results
The expression levels of wild-type ERP-2 were disappointing at 0.5 mg/L the IRP2 mutants in baculovirus (10 mg amounts needed for crystallization), but there was hope
that without the degradation domain IRP-2 would express better. The mutant IRP-2s
didn’t express any better than the wild-type IRP-2 when western blots were compared at
the primary virus isolate stage. The expression of the IRP-2 mutants in baculovirus was
not pursued further. Later on, George Gauss and Philippe Benas were able to express
wild-type IRP-2 (Img/L) and IRP-I (2-3mg/L) respectively in yeast cells (saccaromyces
ceriviciae). Preliminary expression trials of IRP-2-ADD indicated that it expressed at
0.5mg/L, further expression trials will need to be performed in order to determine if
expression of the IRP-2 mutants should be pursued further.
33
CHAPTER 3
Synthesis and Purification of IRE
IRE Construction
The main goal of the Lawrence Lab is to crystallize IRP/IRE complex for
crystallographic studies. Therefore, two things that are required to accomplish this goal
are purified IRP and purified IRE. IREs of varying sequence are found in many genes
regulated by IRP-I and IRP-2, and these sequences also vary between species. Also,
there are two major forms of IRE; the IREs for ferritin synthesis (Bulge/Loop IRE) and
the IREs for transferrin receptor (c Bulge IRE) (38). Thus, there is great diversity in IRE
structure leading to many possible choices in the IRE to be expressed for structural
studies. Fortunately, IRP-I shows broad specificity and binds to a variety of IREs equally
well, while IRP-2 seems to be more specific (38). Therefore, a ferritin sequence of 30
nucleotides from rat was chosen because it can bind to both IRP-I and IRP-2 (Fig. 17).
a
A
c
u
G
c
A• U
A• U
C• G
U• A
U• A
C
G
C
U
C•
C•
U•
U•
G
G
A
A
U• C
I RE
Fig. 17: Wild type ferritin IRE sequence from rat.
34
Two common routes for production of small RNA molecules are chemical
synthesis and enzymatic synthesis. Chemical synthesis is usually done commercially; the
desired sequence is simply purchased from a company such as Dharmacon. In contrast,
enzymatic synthesis using T7 RNA polymerase is more easily done “in house”. An
important consideration is the need for milligram amounts of homogenous IRE that is
necessary for crystallization trials.
We initially purchased IRE from an RNA synthesis company (Dharmacon). The
requested sequence for the IRE: (5’ UUUCCUGCUUCAACAGUGCUUGAACGGAAC
3’) was sent into Dharmacon, and for $374 the lab received ZOOnmoIes (0.79 mg) of IRE.
This IRE was primarily used for early binding assays with IRP-2, and later used to check
the size of the homemade IRE in denaturing PAGE gels. From these experiments it was
determined that the IRE was not a homogeneous 30 nucleotide sequence, but consisted of
at least two fragments, one at 30 nucleotides and the other at 60 nucleotides (Fig. 18).
Multiple
Bands of
Synthetic
IRE
Increasing Concentration
Fig. 18: Synthetic IRE appears to have multiple bands; the two most prominent are a
30mer and a 60mer. Synthetic RNA was loaded with increasing concentrations on a
denaturing PAGE gel that was ethidium stained and checked with a UV light.
35
Considering the cost for milligram amounts of IRE and the heterogeneous mixture of IRE
provided by Dharmacon, it was determined that we should undertake enzymatic synthesis
of the IRE.
Two different strategies were used to make homemade IRE in the laboratory; both
involve using in vitro T l RNA polymerase transcription reactions. One is to amplify
desired template by PCR or use annealed oligos containing a T l promoter to generate a
run-off 30 nucleotide sequence of the rat ferritin IRE, but this will result in heterogeneous
ends (39). Typically, the T l RNA polymerase can add anywhere form one to three extra
nucleotides to the 3’ end of the desired sequence, this can cause heterogeneity in the
3’end of any RNA synthesized by T l RNA polymerase (39). The other method is to
design hammerhead ribozymes, from the group one intron in Tetrahymena thermophila,
on either side of the IRE to precisely cleave the IRE to leave homogenous ends (40).
The self-splicing pre-ribosomal RNA (rRNA) of the ciliate Tetrahymena was the
first example of an RNA molecule that could form a catalytic active site for a series of
precise biochemical reactions (40). The ribozymes in Tetrahymena self-splice the introns
away from the rRNA allowing the exons to join together and be translated (41). The
discovery of ribozymes increased the credibility of the RNA world hypothesis, where
RNA served both as the genetic material and the principal cellular enzyme (41).
Descendants from this proposed RNA-dominated world occur in the form of natural
ribozymes present in organisms ranging from bacteria to humans (41).
Hammerhead ribozymes (HHR) are the smallest of the naturally occurring
ribozymes (~40 nucleotides), and recent experiments show that the hammerhead motif is
36
the most efficient self-cleaving sequences that can be isolated from randomized pools of
RNA (41). The HHR catalyses site-specific cleavage of one of its own phosphodiester
bonds via nucleophilic attack of the adjacent 2’-oxygen at the scissile phosphate (Fig. 19)
(41). Initial studies revealed a requirement for a divalent metal ion for catalysis (Mg2+) to
stabilize the active HHR structure, but its direct role in RNA cleavage is unknown (41).
transition stale
Fig. 19: The proposed HHR cleavage mechanism. The 2’-oxygen atom attacks the 3’phosphorus group with expulsion of the 5’-oxygen atom proceeding via the trigonal
bipyramidal transition state. The reaction could be accelerated by removal of the proton
from the 2’-oxygen atom by a base (shown as B-) and protonation of the departing
oxyanion (by B H).
The first strategy (run-off IRE transcripts) will generate some heterogeneity in 3’
ends of the IRE, but it should not affect IRP binding since the extra nucleotides are
distant from the more critical hexa loop and loop/bulge region of the IRE. However, the
extra nucleotides may interfere with crystal packing if they protrude too far out of the
complex. The second strategy utilizes HHR to yield homogenous IRE with defined 5'and 3'-termini. Upon addition of Mg2+, the HHR catalyzes precise cleavage reactions
excising the IRE (Fig. 20).
37
5’ HHR
IRE
3’ HHR
cT r T 3
HHR cleavage on 3’ end of C
Hind 3
UU
C
CUGACCG
G
I R E
GACCCUGCUU
GACUGGCA
AA
C A A C A G
CUGCAG
AC
A
G
C
C
C
UG
A
U
C
C
AG
C
A
G
G
U G C U U G A A C G G G Ul
•
•
•
•
G
G
U
C
C
A
G
G
Xho
I
RNA
C C G A G C U C G C C G G C 5 '
G
G
U
Fig. 20: The HHR-IRE-HHR construct. Visual representation of what the construct
looks like (above), as well as the sequence of the 5’ HHR. When the 3’and 5’ HHR fold
in the presence of Mg2"1"they will precisely cleave away leaving the IRE on the 3’ end of
the cytosine (black box). Red text indicates IRE sequence, green text indicates 5’ HHR
sequence, blue text indicates restriction sites, and grey boxes indicates base pairing.
Philippe Benas (Montana State University) designed the sequence for two HHR
(Fig. 20) and then cloned the sequence for two HHRs separated by a Sal I cloning site
into the pBluescript vector, which was named pBPBE-3. The cloned sequence for rat H-
38
ferritin IRE was inserted into the Sal I site. The Sal I site allows flexibility to change the
IRE.
For HHR cleavage, it was necessary to modify the IRE lower stem sequence to
allow base pairing with the HHR (Fig. 21). This sequence change should not affect IRP2 binding because of the introduced compensatory sequence changes to maintain full base
pairing of the lower stem.
G
A
C
A
A
C
U
U
•
•
•
•
•
U
G
C
U
U
G
A
A
C
G
C
U
C
C
C
A
G
•
•
•
•
•
G
G
G
U
C
I RE
Fig. 21: Mutant ferritin IRE that varies from the wild type ferritin IRE in the lower stem
to allow base paring with the HHR.
By using a combination of the two strategies (PCR amplification of desired
template, HHR) there are many different possible reactions that can be run to synthesize a
variety of RNAs (Fig 22). By using different primer combinations with and without the
pBPBE-3 plasmid a variety of template DNA can be made such as: wild-type IRE, a
mutant ERE, 5’ HHR-IRE-HHR 3’, IRE-HHR 3’, and plasmid sequence-5’ HHR-IREHHR 3’-plasmid sequence.
39
Because the pBPBE-3 was designed to include restriction sites within the HHRIRE-HHR sequence, ferritin IRE can be changed to a transferrin receptor IRE, other
IREs, or different HHR sequences can easily be swapped in allowing further choices in
IRE and HHR constructs.
DNA template for 17 Rxns
IP2-RIR
BP37-RIR
HHR-IRE-HHR
IIPS-RIR
BP37-RIR
IRE-HHR 3’
RunoffIRE
Fig. 22: The pBPBE-3 vector and various primers can be used to generate a
variety of RNA fragments including: 5’HHR-IRE-HHR3\ IRE-HHR3’, and run­
off IRE (wild type and mutant). Restriction sites are also shown.
40
RNase Control
RNases are a family of enzymes that degrade RNA rapidly by cleaving the RNA
at specific sites, without RNases the amounts of mRNA in an organism would
accumulate and cause problems with the protein expression levels in the organism.
However, it is essential to inhibit these RNases if you would like to work with them in
the laboratory environment because the RNases will degrade the RNA before any
meaningful experiments can be completed. Controlling and monitoring RNase activity
are essential techniques that must be learned in order to work with RNA. Common
sources for RNase contamination and ways to control it are (42):
1. "Fingerases"; bodily fluids, and skin cells are often referred to as "fingerases"
2. Tips and tubes; autoclaving alone will not destroy all RNase activity; these
enzymes are very robust and can refold upon cooling to room temperature to
regain partial activity.
3. Water and buffer; DEPC-treated water is the most common method used to
inactivate RNases.
4. Laboratory surfaces; benchtops, glassware, metalware, plasticware, and other
surfaces that are exposed to the environment; glassware and metalware can be
baked overnight at 450°F to inactivate RNases.
5. Endogenous RNases; all tissue samples contain endogenous RNases.
6. RNA samples; small amounts of RNases may co-purify with isolated RNA.
41
7. Plasmid DNA; DNA used for in vitro transcription and other reactions can
introduce RNase contamination, a phenol/chloroform extraction can be used
on the DNA to help eliminate RNases.
8. RNA storage; to preserve isolated RNA for long-term storage a salt/alcohol
precipitation is essential; typically the nucleic acid is stored as a precipitate in
solution.
9. Chemical cleavage; RNA molecules can also undergo strand scission when
heated in the presence of divalent cations such as Mg2"1", Ca2"1",and Pb2"1"at
>80°C for 5 minutes or more' ■
10. Enzymes; both commercially purchased and laboratory prepared enzymes.
RNases can be inhibited though a variety of methods, some work better than
others, while others are easier to use. Strongly acidic (pH<4) or basic solutions (pH >9)
will denature the RNases and are useful to clean pH electrodes, UV-cells, and
purification columns. Hydrogen peroxide (H2O2) at a concentration of 3% (or more) for
ten minutes causes an irreversible inhibition of the RNase and is easily removed with
several washes of RNase-Free water; the use of H2O2is practical for cleaning glassware.
The addition of 5mM EDTA will cause reversible inhibition by chelation of magnesium
(Mg2"1"), magnesium is an essential component for optimal RNase activity (and for folding
of the HHR), it is useful for the long term storage of RNA but EDTA can be difficult to
remove completely. DEPC at 0.05% causes irreversible inhibition it is commonly used to
treat water and to clean glassware, but it can be difficult to remove and can react with
42
nucleic acids (causes the aromatic rings to open). Denaturing agents and detergents are
also used but are difficult to remove later on. Commonly used denaturing agents are; 10%
SDS, Urea (3M), and Guanadinium chloride (6 M). Commonly used detergents are 1%
Triton X-100, and 1% Tween 20. There are also many commercially available RNase
eliminators to clean glassware; RNase Zap (contents unknown) by Ambion is frequently
used in our lab. RNase Zap is easy to use, apply a few drops to the desired piece of
glassware, wipe it down with a lint free paper towel, and wash it thoroughly with RNaseFree water. A non-chemical alternative to inhibiting RNases wash the glassware with
five volumes of water and then bake the glassware at 450°F for six to eight hours or
more. Autoclaving the glassware will not destroy RNases. Metal instruments can be
cleaned in a similar way, by heating the metal instrument to high temp with a Bunsen
burner and rinsing it with RNase Free water. RNase activity also needs to be controlled
at the reaction stage for RNA transcription by adding specific RNase inhibitors
(RNasins). I Unit/ml RNasin inhibits RNase A, while I Unit/ml SUPERaseoIn (Ambion)
inhibits RNase A, RNase T l, and RNase C. Ethanol (95%-100%) was used to clean the
bench top, gloves, the outside of columns and tubing, and pipetmen. Table 2 lists some
common enzymes that can modify nucleic acids (43).
Table 2: Nucleic Acid modifying enzymes
RNase H
RNase
Endoribo
RNase A
RNase
Endoribo
RNase Tl
RNase U2
RNase
RNase
Endoribo
Endoribo
RNase CI-3
RNase
Endoribo
RNase S7
RNase <i)M
DNase/RNase Endonuch
RNase
Endoribo
RNase
Endoribo
RNase T2
DNase/RNase Endonucl.
Nuclase Pl
DNase
Endonucl.
DNase I
Snake venom
Exonucl.
Phosphodiesterase
Spleen
Exonucl.
Phosphodi esterase
Alkaline
Phosphatase (PAL)
Cleaves RNA
5'p-RNA
only in DNA-RNA hybrids
Cp/N
RNA-p3'
Up/N
Gp/N
RNA-p3'
Ap/N
RNA-p3'
(Gp/N)
Cp/N
RNA-p3'
(Ap/N)
(Gp/N)
Np/A
(D/R)NA-p3'
Np/U
Np/U
RNA-p3'
Np/A
Np/G
Np/N
RNA-p3' Unspecific endoribonuclease
Np/N
(D/R)NA-p3'
N/pN
DNA-p3' Unspecific endodeoxyribonuclease
Unspecific 3' - 5' exonuclease
N/pN
RNA-0H3'
requires RNA with free 3'0H
Unspecific 5' - 3' exonuclease
Np/N
RNA-p3'
requires RNA with free 5'0H
p/N
5'OH-N
Removal of all terminal phosphate
groups
Table 2: continued: Nucleic Acids modifying enzymes
Polynucleotide
kinase (PNK)
PNK
3' phosphatase
mutant
pol
p-p-p-rA + 5'HO-poly(N)
5'p-poly(N)
pol
ibid
RNA ligase
pol
PolyA polymerase
pol
Terminal
deoxynucleotidyl
Transferase
T7 RNA polymerase
(T7RNP)
RNA30H + pppORN
RNA~pORN
RNA3 0H + n (d/r)ATP
RNA~(pA)n
5'p-(D/R)NA Transfer the y-Phosphate of ATP
to the 5 -OH of a nucleic acid
ibid
Adds 5'-phosphate donor RNA
to the 3 -OH ofRNA
Adds ATP (r or d) to the 3 -OH of
RNA-(pA)n
RNA
RNA-ORN
pol
DNA3 0H + n dNTP
DNA~(pN)n
pol
DNA-dep RNA pol
RNA
RNA-dep DNA pol
DNA-dep DNA pol
DNA
RNase H
domain
DNA-(pN)n Adds dNTPs at the 3' of DNA
Reverse transcriptase
(RT)
pol
Taq pol
pol
DNA-dep DNA pol
DNA
Vent pol
pol
DNA-dep DNA pol
DNA
PPi -» 2Pi
Pi
Pyrophosphatase
(PPiase)
Apyrase
ppp-rA
2 Pi + p-rA
Useful for 3' end-labeling
rAMP
Uses DNA as template, rNTPs as
primers
Uses either DNA or RNA as
template
Uses either DNA or RNA as
primer
Uses DNA as template, DNA as
primer
Uses DNA as template, DNA as
primer
Also removes PPi groups in cap
structures
Hydrolysis of the pyrophosphate
bonds in rATP
45
For assembly of the IRP/IRE complex, the RNA must be correctly folded. Thus
our protocols must use reagents that can be easily removed and will not disrupt the RNA
secondary structure. Therefore, 30% H2O2and RNase Zap were commonly used to clean
glassware. 30% H2O2is added to the glassware/plastic ware with shaking for ten minutes
(or let stand for thirty minutes) then remove the H2O2(can be reused for at least one
month before the H2O2goes bad) and wash with five volumes of RNase-Free water.
RNase Zap was only used when glassware needed to be cleaned right away, since it is
much faster, wipe on and wash off (five volumes of RNase-Free water). Metal
instruments were cleaned as described previously, and RNasins were used at the
concentration mentioned above as well.
Un-opened boxes of gloves were assumed to
be RNase free, and great care was taken when putting on the gloves to ensure that
potential “fingerases” didn’t contaminant the exterior of the gloves. When in doubt,
potential contaminated gloves were exchanged for clean gloves. Ethanol (95%) was
commonly used to clean gloves and pipetmen at any stage to ensure good sterile
technique and to wash off any potential RNases.
RNase-Free solutions are needed for the transcription reactions, and purification
steps. The reagents necessary to make the solutions were purchased separately and kept
isolated from general lab use. Only people currently working with RNA are allowed to
use these reagents. RNase-Free bottles were used, and unopened Falcon tubes (50ml and
15ml tubes, assumed to be RNase-Free) to mix and store the solutions. The powdered or
liquid reagents would be added directly to the bottle (without a spatula), and by placing
the bottle on scale it would eliminate the use of a weigh boat or weigh paper. This
46
eliminated any chance of RNase contamination coming from weighing out the reagents.
To adjust the pH of the solutions, an RNase free pH probe would be used, or two
identical solutions would be made where one of them would be checked with a regular
pH probe. By keeping track of how much acid or base that was added to the non-RNase
free solution, the second solution could be made with out the use of a pH probe. To
assure that the pH of the RNase-Free solution was correct, we commonly take a small
aliquot of the solution and check it with a pH meter.
It is necessary at times to check solutions for RNase activity. This includes all
stock solutions, transcription reactions, and subsequent purification steps, so that RNA
degradation will not interfere with complex formation or crystallization. One way to do
this was with a simple time course, just add some RNA to a tube containing the solution
that needed to be check and remove small aliquots and run them out on a denaturing
PAGE gel. The gels proved difficult to read, so a more quantitative method was needed
to detect for RNases. A more effective and quantitative way to detect for RNase activity
is to use Ambion's RNAseAlert kit. The kit contains an RNA oligo that is double labeled
with fluorescent and quenching moieties, when RNases are present they open up the
RNA oligo allowing the fluorescent marker to be excited at the proper wavelength of
light. The excitation wavelength is 490nm, and the emission wavelength is 520nm. The
procedure is simple and allows for multiple samples to be detected using a fluorescent
plate reader. Sample aliquots of 45
had RNA oligo added, placed into a plate reader
and readings taken every five minutes for one hour (Fig. 23).
B uorescent
Substrate
V
•R e su sp e n d d r y F lu o re sc e n t S u b stra te
w ith B pi R N a se A Ie rt Buffer
•A dd u p t o 4B pi te s t sa m p le
Incubate assay
3 0-60 m in s .a t 37 C
D e te c t flu o re s c e n c e
Uanc ° handheld UV Iieht or a transilluminator
or
Udne a Hoorometer ip late o r tube.1
A
/ \
\
\
V
R N a s e -c o tita m in a te d
S o lu tio n
R N a se -F re e
S o lu tio n
Fig.23: Ambion’s RNase Alert Test kit procedure. When RNases are present the quench
will be cleaved allowing the substrate to fluoresce.
A Tecan-SAFlRE absorbance and fluorescence plate reader (Edward Dratz, Montana
State University) was used to read the samples. Experiments indicated that a 5x dilution
of the RNA oligo could still generate a strong signal in the plate reader from an RNase
positive control. The kit was used to check all solutions, transcription reactions, and
fractions from purification steps. The kit also confirmed that our milliQ water from our
water purifier was RNase-Free, so no further water treatments had to be used to ensure
that our water was RNase-Free. It is important to note that if you want potential RNases
to be fully active to detect with the kit, there should be 5 mM MgCla and I mM DTT in
the medium (these components are included in the RNase Alert 10 x buffer).
48
T7 RNA Polymerase Expression and Purification
Introduction
The interest of the lab is to eventually obtain x-ray crystallographic structures for
the IRP-1/IKE and IRP-2/IRE complexes. Making these complexes will require a great
deal of RNA to be produced. The most expensive of the reagents necessary for a T7
RNA polymerase reaction is the enzyme, T7 RNA polymerase (T7RNP), so it was
critical to be able to make homemade T7 RNA polymerase in the lab to cut costs.
Elizabeth Leibold at the University of Utah graciously donated a plasmid DNA
containing the sequence for the T7 RNA polymerase, flanked by a sequence coding for a
6xHis tag, and a detailed protocol for the expression and purification of the T7 RNA
polymerase.
Expression of the T7 RNA Polymerase
E. coli (BL-21 strain) were transformed with the plasmid DNA and plated on LB
ampicillian (lOOpg/ml) plates. A colony containing the T7RNP was picked and grown in
overnight starter culture at 37°C in LB media (50ml) with ampicillian (lOOpg/ml). The
starter culture was then transferred into 2 liters of terrific broth (12 g Bacto tryptone, 24 g
Bacto yeast extract, and 4 ml glycerol per 900 ml; autoclave solution, then add 100ml of
filter sterilized 0.17M KH2PO4 and 0.72M K2HPO4) at 37°C and grown to an O.D.goo of
0.4. IPTG (IOOpM) was added to the culture to induce protein expression. Protein
expression continued for 3 hours at 37°C with shaking. The cells were then pelleted and
resuspended in 30 ml of lysis buffer (50mM Tris-HCL pH 8.0, IOOmM NaCl, 5mM P-
49
mercaptoethanol, 5% glycerol, and ImM imidizole). Cells were lysed using a
microfluidizer (Valerie Copie, Montana State University), and spun in an ultra centrifuge
(45,000 RPM for 2 hours) to remove cellular debris.
Purification of the T7 RNA Polymerase
The supernatant was passed over a Ni-NTA affinity column (equilibrated with
lysis buffer). The Ni-NTA resin (40 ml) was washed with lysis buffer 3x40 ml, followed
by a second washing with lysis buffer plus IOmM imidizole 3x40 ml. The T7 RNA
polymerase was eluted with lysis buffer plus IOOmM imidizole 2x20 ml. The
concentration of the T7RNP was determined by the Bradford assay. The T7 RNA
polymerase was dialyzed against storage buffer 8x500 ml (IOmM Tris-HCL pH 8.0,
ImM EDTA, 50% glycerol, IOOmM NaCl, and ImM DTT), exchanging the buffer
approximately every two hours. A final concentration for T7RNP was obtained, and
checked for RNase activity using RNase Alert assay. The T7 RNA polymerase was
aliquoted into 1.5 ml microfuge tubes, and frozen away at -20°C. All buffers were
RNase Free, as determined by the RNase Alert Assay.
T7 activity assay
Once the enzyme was purified, the activity of the T7RNP needed to be
determined. Whenever fresh T7RNP is purified the specific activity and the total activity
of the purified T7RNP should be measured, along with the concentration and the total
amount of the protein. Quantification of radiolabeled transcripts produced by in vitro T7
50
transcription is used to measure the enzymatic specific activity of home-purified T7 RNA
polymerase. The quantification is based on the selective precipitation of the radiolabeled
transcripts on cellulose membrane by trichloroacetic acid (TCA) (44).
Like all polymerases, T7RNP uses nucleotides triphosphate (NTPs) as substrate
as well as Mg2"1"and a template DNA as cofactors to synthesize oligonucleotides.
Polymerases can be classified following the nature of the neo-synthesized oligonucleotide
and the nature of the template used. Based on this classification, T7RNP is a DNAdependent RNA polymerase that uses ribonucleotides triphosphate (NTPs; N= A, U, G,
C) as substrate to produce RNA by the following scheme (Fig. 24) (45):
T7P-Templaten/T7RNP
RNAi + rNTP
t
RNAm + PP1
fori = Oton
T7P = T7 Promoter
n
= number of template nucleotides
RNAj = neo synthesized RNA containing I
nucleotides
PPi
= inorganic Pyrophosphate
Fig. 24: T7 transcription reaction scheme
The T7RNP specific activity assay is based on the time course quantification of
transcripts synthesized. This quantification is performed by selectively precipitating the
oligoribonucleotides on cellulose membrane and by determining the number of
radiolabeled UMP incorporated (44, 45). This approach has three major advantages:
I. The use of radioactivity provides a precise quantification applicable at analytical
scale, hence minimizing the amounts of T7RNP and NTPs needed.
51
2. Radiolabeled transcripts, of length equal or greater than 15 nt are precipitated by
trichloroacetic acid (TCA) on the cellulose membrane (Whatman 3MM paper),
whereas uncorporated NTPs are not retained.
3. An answer can be obtained within 3 hours.
In all experiments, pyrophosphatase (PPiase) was added to shift the equilibrium
towards product formation; this enzyme catalyzes the hydrolysis of pyrophosphate (PPi)
into inorganic phosphate (Pi).
Methods
When preparing a working solution of radiolabeled XTP, the specific
radioactivity ( S A Xt p , work) of the given radiolabeled XTP* is required for measurements
on scintillation counter efficiency (E for the given isotope).
SA
xtp,
work
can be determined
by the following relation (45):
CpiU assay
SA XTP, work (dpm/mol XTP) =
___________________
Eq. (I)
Ymin [DNA] Vassay E Nx
Where cpmassay represents the number of expected counts per minute (cpm), Vassay
is the volume, [DNA] is the DNA template concentration necessary to synthesize at least
Y m in- f o l d
more transcripts during the minimal interval time, and the full-length products
containing Nx nucleosides (with X being of the same nature as the XTP* used in the
reaction) (45).
For example, if the efficiency of the counter is 40% (takes also into account all
quenching phenomena), if the yield at 5 min. (shortest time point) is 1.4 full-length RNA
52
(containing 121 U) per molecule of template DNA (present at 50 nM in the reaction), and
if 2000 cpm are wanted in a 15 pL assay, then:
2000
SA XTP, work =
----------------------------------------------------------------------
1.4 * 50.IO"9 * 15.IO"6 * 0. 4 * 121
= 39.3.IO12 dpm/mol of [a-35S]-UTP
i.e. SA XTP, work = 39.3.IO12/ 2.22.IO12 = 17.7 Ci/mol
The only assumption made in this relation is that all products are full-length,
which is obviously wrong, but sufficient to calculate the useful isotopic dilution D, given
by:
D j = S A XTP, w ork / S A XTP, stock
Thus, for a [a-35S]-UTP stock solution is at S A
X t p , stock
= 400 Ci/mmol, then the
isotopic dilution is: D; = 17.7/400. IO3 = 1/22587.
A Vwork pL working stock solution of radioactive UTP, at a concentration CW
Ork,
can then be made according to the following equations system:
("C-XTP, stock * VxTP, stock * S A x T P 1stock — U w o rk * V work * S A XTP, w ork
(1)
Eq. (2)
IC x T P , stock * VxTP, stock + CxTP, cold * VxTP, cold = C work * V w o rk
( 2)
Where C x t p , cold and V X t p , cold are the concentration and volume of non-radioactive
XTP to mix with the radiolabeled XTP (45).
(I)
= > V xTP, stock — C w o rk * V w o r k
CxTP. stock
*
S A XTP, work — C w o rk * V w ork * D j
S A xT P , stock
CxTP, stock
Eq. (3)
53
(2) => V x T P , cold
— [ C w o rk * V w o r k ~ Q xTP, stock
* ^ X T P , stock ]
CxTP, cold
Eq. (4)
For example, to prepare 22.6 pL of [a-35S]-UTP working solution at SA x t p , stock =
17.7 Ci/mol and 50 mM, using a 100 mM stock solution of non-radioactive UTP, and a
stock solution of [a-35S]-UTP at 400 Ci/mmol, with a total activity of 250 JJ-Ci in 25 pL,
we have (45):
250.10'6/400.IO3 = 625.IO"12 mol of [a-35S]-UTP, i.e. 625 pmol, in 25 pL
=> C
x tp,
stock
= 625.IO'12/ 25.10"6 = 25. IO"6 M, i.e. 25 pM
Using Eq. (3):
V x t p , stock = 50. IO"3 * 22.6 * 17.7
= 2.0 jxL
25. IO"6
400. IO3
And Eq. (4): V X t p . cold = 50. IO'3 * 22.6. IO"6 - 25.10"6 * 2. IO'6 = 11.3. IO"6 L
100. IO"3
i.e. 11.3 JtL of non-radiolabeled UTP at a concentration 100 mM should be mixed
with 2 JtL of the [a-35S]-UTP stock solution (250jtCi) and adjusted to a final volume of
22.6 JtL (using a 40 mM Tris-HCl pH 8.1 solution). This working solution can be stored
at -20°C.
Preparation of template DNA must be taken into account when performing this
assay. In order to avoid premature termination of the T7 transcription due to the
particular sequences in the IRE constructs, and hence have a standard protocol for the
T7RNP activity assay, it was decided to use a plasmid containing a T7 promoter as
template: pBluescript //SK+ (46). Although not required to determine the T7RNP
54
activity, knowledge of controlling the maximum number of U incorporated in the fulllength transcript is necessary to estimate the T l transcription reaction yield. The T l
transcription yield is defined as the number of full-length RNA molecules made per
template DNA molecule, which is the parameter used in finding optimal conditions for
the reaction. To avoid “run-around” transcripts and produce transcripts of defined length,
the plasmid is linearized by Afl III, a specific endonuclease. This restriction enzyme
hydrolyzes the chosen plasmid once, after the cytosine in the following recognition
sequence: (+) DNA 5’ .. .C/TTAAG .. .3’, 508 nucleotides downstream the G+i of the T l
promoter, so that the number of U incorporated in a full-length RNA is 121.
Assay conditions are under which T7RNP is limiting (pseudo first order kinetics) and
having adjusted the [a-35S]-UTP specific activity as required. The following protocol for
T7 transcription and TCA precipitation/wash was designed and optimized by Philippe
Benas and myself (45):
1. Use one reaction per time point, in order to avoid concentration uncertainties that
could occur if aliquots were taken (in particular if precipitation of the transcripts
occurs as more RNA is made)
2. Use a minimum reaction volume of 17 pL to minimize evaporation effect on
concentrations. This allows one to remove a constant and precise volume of 15 pL
for each sample
3. The 17 |±L reaction mixtures contain 50 nM linearizedpBluescript IISK.+, 40
mM Tris-HCl pH 8.1 at 37°C, 0.01% (v/v) Triton X-100, 50 mM DTT, 8% (v/v)
55
PEG 8000, 50 |a,g/ml BSA, 400 U/mL RNasin, 2.5 mM NTPs including [a-35S]UTP at a specific activity of 17.7 Ci/mol (prepared as described above), I U/mL
PPiase, 16 mM MgCla, 2 mM sperm dine and the T7RNP. The addition order is as
listed and the T7RNP addition corresponds to t = 0. All solution but the T7RNP
are brought to room temperature before addition to the reaction, spermidine and
MgCla being known to precipitate DNA at colder temperature (e.g. 4°C).
4. Reactions are carried out at 37°C, and 3 time points of 15 pL each are taken that
varied depending on the assay from: 2.5, 5, 10, 15, and 20 min.
5. The T7RNP concentrations were the following: 233 nM for a commercially
available T7RNP (TaKaRa), 207 nM T7RNP-2 (10 months old, stored in 50%
glycerol), 5.8 jrM T7RNP-3, and for 4.6 pM T7RNP-5.
6. One positive control is to deposit 15 pL out of 17 pL rxn. (contains all
components) on 3 MM paper, that is not followed by a TCA precipitation or any
wash. Ideally, there should be one positive control per time point, but in practice,
a single positive control is sufficient. This simulates the maximum number of
counts that could be incorporated into the full-length RNA transcripts.
7. A measurement of background, a 15 pL blank, contains all the components of the
T7 transcription and 5% TCA, prior to T7RNP addition; several blanks are
required for a good estimation of the background (e.g. one per time point); no
correlation with incubation time was ever observed.
To count the amount of radioactivity, I cm x I cm cellulose 3MM paper
(Whatman) pieces are previously pinned on a piece of Styrofoam. 15 pL out of the 17 pL
56
reaction mixture are pipetted onto a 3 MM paper, air-dried for 5 seconds and then
dumped into 5% TCA, chilled by melting-ice (O0C). The TCA stops the reaction by
denaturing the RNA polymerase and precipitates the oligonucleotides longer than 15 nt.
At the end of the experiment, two additional washes of 10 min. each in chilled 5% TCA
are carried out, followed by 3 washes of 10 min. each in 95% ethanol. The pieces of
3MM paper are put back on the Styrofoam for drying (either under a heating lamp for at
least I hour or overnight at room temperature). Finally, each piece of paper is placed in a
scintillation vial containing ImL of scintillation liquid (Ecoscint 0), and counted for 10
min. on a Packar 2200 CA Tri-Carb counter.
Results
The measurements obtained from the counter and initial speeds are summarized in
Table 3, whereas Figures 25, 26, and 27 give a graphical representation the number of
nmoles of [a-35S]-UMP incorporated in transcripts as a function of reaction time.
5
10
15
20
578.30
91.20
286.80
272.50
608.66
95.98
301.99
286.78
61.80
71.66
65.28
64.64
10.00
10.00
10.00
10.00
5
10
15
20
3539.31
7677.55
10285.70
3633.55
3726.14
8086.77
10826.30
3831.90
60.36
57.81
56.82
58.09
5.75
2.65
1.42
1.28
Table 3: CPM and DPM counts for blanks and T7RNP preps, and initial rates of
T7RNP preps
57
5
10
15
20
2151.64
2415.52
2823.93
4552.85
2268.31
2540.70
2974.88
4794.03
62.60
59.18
58.77
59.09
5.17
4.64
3.97
2.46
5
10
15
20
23358.30
45632.00
74506.30
27685.40
24619.40
48062.60
78439.30
29181.10
58.13
64.56
60.15
57.92
0.48
0.25
0.16
0.41
10
1065100.00
1100400.00
1208239.00
1253542.00
22.41
22.14
0.02
0.02
2.5
5
10
20
9320.78
10469.90
18731.20
202600.00
10617.30
11911.30
21310.80
231002.00
24.69
26.03
24.38
24.28
1.54
1.36
0.77
0.70
2.5
5
10
20
14007.80
14788.70
46546.90
287067.00
15870.00
16818.20
52872.40
326580.00
24.54
24.62
23.12
23.97
1.02
0.97
0.32
0.60
5
Speed
(nmol/min)
4.32
29.43
N/A
220.09
1.46
0.86
11.00
0.26
19.07
Time
Speed (nmol/min)
Speed (nmol/min)
Speed (nmol/min)
5
10
15
20
Speed (t=5)
Speed (t=20)
Speed 5-2.5
Speed 20-10
Average DiVDt
Average n/t
Average 20 min
5.12
11.67
15.79
5.28
1.02
1.17
1.05
1.31
0.82
1.07
1.08
2.92
3.33
3.99
6.72
0.58
0.33
0.27
0.08
0.13
0.11
0.39
36.53
71.78
117.45
43.39
7.31
7.18
7.83
7.05
9.13
8.09
7.44
A S (nm ol/hour)
N units//il
1.08
140.30
0.28
40.87
7.70
40.42
15.03
238.32
N units/ml
2.50
0.73
0.72
4.25
Table 3: continued
58
c
700
Commercial T7RNP
600
1st T7RNP
y = 47.096X + 8.1515
R2 = 0.9829
3L 400
S
300
O 200
4.6868X + 103.79
R2 = 0.9828
time (min)
Fig. 25: [a-35S]-UTP incorporation using the commercial or T7RNP-1
Fig. 26:
[Ct-35S ]-UTP
incorporation using T7RNP-3
59
250000
d 200000 -
•
T7RNP-5
=? 150000
o 100000
I
y = 14468x - 70508
R2 = 0.9991
50000
time (min)
Fig 27: [cc-35S]-UTP incorporation using T7RNP-5
Although there are only three time points per experiment they are linear, indicating that
the initial reaction velocity is linear over time and 1st order in enzyme concentration. In
addition, the enzymes concentrations used and the products formed are less than 10-fold
and 5% of the initial NTPs concentration respectively, which corresponds to conditions
required for kinetic analysis of order I. Table 4 gives the relative specific activities for
the three T7 RNA-polymerases used in the present experiments.
Table 4: Relative specific activities of the T7RNP
Relative
Specific
Activity
100%
29.2%
28.8%
170%
60
Discussion and conclusions
The purpose of the experiment was to provide milligram amounts of active T7RNP at an
affordable price comparable with the quality of the commercially available T7RNP.
Purifications typically yielded 40 mg/L of T7RNP per liter of E. coli. The activity assay
indicated that the T7RNP-2 and T7RNP-3 purifications had 3 fold lower specific
enzymatic activity compared to the commercial one, while the latest prep (T7RNP-5) is
about 2 fold more active than the commercial T7RNP. Hence, our purification scheme is
rather well designed. In addition to determining the enzyme activity, the T7RNP was
tested for RNase activity using the RNase Alert kit. Earlier enzyme purifications
(T7RNP-2 and -3) contained a significant amount of RNase (~ 0.5 pg/ml), this possibly
accounts for the lowered specific activities. The RNase contamination was first
attributed to insufficient stringency during elution at the affinity chromatography step
(Ni-NTA) and/or possible contamination during dialysis. Therefore, in the following
purification (T7RNP-5), all solvent components were checked for possible RNase
contamination before use by using the RNase Alert. The RNase assay still detected
RNase activity from the eluted protein, but with extensive dialysis with RNase-Free
storage buffer (8 washes of 500ml each), the T7 RNA polymerase showed no RNase
activity. This suggested that it was not necessarily critical to have RNase-Free solutions
during the column washes and elution. Of paramount importance is extensive dialysis
against RNase free storage buffer, at least 8x500ml buffer exchange at 2 hour intervals.
Obtaining milligram amounts of T7RNP that was 2-fold more active than that of the
commercial polymerase and RNase free was critical for large scale synthesis of RNA.
61
PCR Amplification of Template DNA
In order to synthesize large amounts of RNA, there needed to be a ready supply of
DNA template for the T l transcription reactions. There are two ways to make DNA for
T7 transcription reactions; one is to linearize a plasmid containing the T7 promoter
followed by your sequence. The other is to use a PCR reaction to selectively amplify your
sequence. The T7 promoter is either carried by the plasmid (with your sequence
immediately following the T7 promoter, unless there is a ribozyme construct) or within
the two primers needed for the PCR reaction. The PCR amplification method was
decided upon because different constructs needed to be tested.
Three PCR templates were amplified; the HHR-IRE-HHR transcription (long
transcript), where the template is a plasmidic DNA (pBPBE3) and one of the two primers
contains the T7 promoter, the IRE-HHR transcription created the same way as the HHRIRE-HHR transcription, and the T7 run off transcription, where the 2 primers are self­
complementary and hence are also template, one of these primers containing the T7
promoter.
Methods
To be able to PCR amplify the plasimd DNA template the annealing temperature
of the primers had to be optimized, this was done by using three different computer
programs. By using Anneal-Hybridize (Olivier Friard, http://annhyb.free.fr). Nearest
Neighbors Method, the melting temperature (Tm) of each primer was calculated. The
62
program takes into account the primer concentration and the sum of the monovalent
cation salt concentrations (80 mM if just the minimal IOx buffer is used).
RNA-Structure 3.5 {David Mathews, http://128.151.176.70/RNAstructure.html)
was used to find possible secondary structures within the primers. The intermolecular
pairing option is used to get both the piimer/primeij and the primeri/primera possible
base pairing.
The data from RNA-Structure 3.5 was used in RNA-Draw {Ole Matzurato,
http://madraw.base8.se) calculate the free energy at various temperatures, and the
temperature that gave a null free energy corresponded to the Tm. The annealing
temperature was obtained by taking the Tm - 7°C of the lowest temperature melting
primer. MgC^ and dNTPs concentrations needed be adjusted between runs.
When performing PCR with primers used as template (self-complementary
primers), the reaction is the same except that the template is made by the hybridization of
the two primers. The primer concentration can be as high as 650 nM, or even more,
provided the Tm of the strongest base paired DNA that could be formed stays reasonable
(max 85°C). A typical PCR reaction procedure is shown in Table 5.
Depending on the type of RNA that was being made the PCR reaction would vary
slightly. The PCR reactions that were optimized were for the wild type IRE, and the
HHR-IRE-HHR constructs. The wild type IRE DNA was formed using BP-25-RIR
primer (5’ GTTCCGTTCAAgcactgttgaagcaggaa 3’) and the BP-34-RIR primer (5’
GAGCGCGCGTAATACGACTCACTATAGttcctgcttcaacagtgc 3’), the bases in lower
case indicates the overlapping sequence.
63
2
PCR "minimum" buffer,
IOx
TrisHCl pH 9.2 @ 25°C
(500 mM)
(NH4)2 SO4 (160 mM)
Triton X-100 (1% v/v)
3
MgCl2 (10 mM)
3
BSA (1500 pg/ml)
3
DMSO (100% v/v)
3
Glycerol (80% v/v)
3
Formamide (100% v/v)
3
Ammonium sulfate
(500 mM)
4
Template DNA (nM)
5
6
Primer 5' (100 [iM)
Primer 3' (100 /xM)
7
dNTPs Mix (IOmM)
8
Taq pol. (5 U//xl)
50 mM final concentration
16 mM in lx solution
0.1 % final cone.
0.5 - 3.5 mM final cone.; Free Mg2+~ 1.5 mM
[MgCl2] = I([dNTPs]) + (0.5 to 2.5) mM
Error rate increases for cone. <1.5 mM of free
Mg2+
10-100 pg/ml final; binds certain PCR
inhibitors
1-10% final cone.
improve the denaturation of GC rich DNA
5-20% final; improve the yield of amplification
1.25-10% final; lowers Tm of melt-resistant
DNAs
15-30 mM final; alters Tm & enzyme activity
0.75 nM final cone.
0.38-1.515 nM final; average = 0.95 nM
Final cone. 156 nM; 100-500 nM final
Final cone. 156 nM; 100-500 nM final
final cone. 200 pM
highest specificity & fidelity at the lowest cone.
final cone. 0.125 U/pl
Pfu pol. @ a final cone, of 0.0125-0.025
U/pL/kb
0.5 pL/20W
Vent pol. (0.01 U//xl)
9
H2O
I
Table 5: PCR reaction using plasmid as template
64
The PCR reaction mixture was assembled using the following reagent conditions:
SOmM Tris pH 9.2, 16mM (NH4)2SO4, 2.3mM MgCl2, 625nM BP-34-RIR, 625nM BP25-RIR, 0.1% Triton X-100, 200pM dNTPS, Taq polymerase 0.05 Units/pJ, and 0.00025
Units/pl Vent. The PCR reaction was cycled under the following conditions: 94°C for
90 seconds, 68°C for 45 seconds, 72°C for 45 seconds, and repeat 10 times; then 94°C for
90 seconds, 80°C for 45 seconds, 12°C for 45 seconds, and repeat 15 times. The DNA
would be checked for purity and the correct base pair length by using PAGE gels (Fig. 28
and Fig. 29).
T 7
run-off
DNA
Fig. 28: Purified wild type IRE DNA generated by BP-25-RIR and BP-34-RIR run out on
a 15% denaturing PAGE gel run at 300V to check its size and purity. Lanes A, B, and C
indicate fractions from a desalting column to remove unincorporated dNTPs. PCR
reactions were purified by performing a phenol/chloroform extraction, isopropanol
precipitation, and run over a desalting column.
65
The HHR-IRE-HHR DNA was formed using the BP-37-RIR primer
(5’ GAGCGCGCGTAATACGACTCACTATAGGGCATGACGTCCTGATGAGTCCG
3’), the BP-22-RIR primer (5’ GGGCCGACGGGTTTCGTCC 3'), and the template
sequence (5 ’CGGCCCGAGCTCGGGCCGACGGGTTTCGTCCTCACGGACTCATCA
GGACGGAGTCTAGACTCCGTCGACCCGTTCAAGCACTGTTGAAGCAGGGTCG
ACGGTCAGAAGCTTCTGACCGTTTCGTCCTCACGGACTCATCAGGACGTCATG
CGGGGGGCCCGGCTCGAGCGGCCG 3’) located in the pBPBE-3 vector.
The PCR reaction mixture was assembled using the following reagent conditions:
50mM Tris pH 9.2 ,16mM (NH4)2SO4, 2mM MgCl2, 156nM BP-37-RIR, 156nM BP-24R1R, 950nM pBPBE-3, 0.1% Triton X-100,125p,M dNTPS, Taq polymerase 0.125
Units/p.1, and 0.00025 Units/p:! Vent. The PCR reaction was cycled under the following
conditions: 94°C for 90 seconds, 63°C for 60 seconds, 72°C for 60 seconds, and repeat
17 times; then 94°C for 90 seconds, 69°C for 60 seconds, 72°C for 60 seconds, and repeat
17 times; then 94°C for 90 seconds, 66°C for 90 seconds, 72°C for 60 seconds, and repeat
10 times. The DNA was checked for purity and the correct base pair length by using
native PAGE gels (Fig. 29).
Other PCR reactions were performed. By using the primer BP-23-RIR (5’
GACCCGTTCAAGCACTGTTGAAGCAGGGT 3’) instead of BP-37-RIR only the
3’HHR would form. By using primers 1224 (5’ CGCCAGGGTTTTCCCAGTCACGAC
3’) and 1233 (5’ AGCGGATAACAATTTCACACAGGA 3Z) instead of BP-37-RIR and
BP-22-RIR, extra plasmidic DNA would be added to the ends of the HHRs.
66
These PCR reaction mixtures were assembled using the following reagent conditions:
50mM Tris pH 9.2, 16mM (NH^SOa, 2mM MgCla, 156nM BP-37-RIR, 156nM BP-24RIR, 950nM pBPBE-3, 0.1% Triton X-100, 125jxM dNTPS, Taq polymerase 0.125
Units/p-l, and 0.00025 Units/pl Vent. The PCR reaction was cycled under various
annealing temperature conditions. The DNA was checked for purity and the correct base
pair length by using native PAGE gels (Fig. 29).
Fig 29: Raw PCR reactions. DNA fragments amplified by primers BP-22-RIR/BP-23RIR (124mer), BP-22-RIR/BP-37-RIR (167mer), 1224/1233 (344mer), BP-25-RIR/BP34-RIR (56mer) are shown on a 15% Native PAGE gel run at 300V.
67
After the PCR was completed the DNA fragments had to be purified to ensure
that it would be RNase-Free and suitable for T7 transcription reaction. Typically the
PCR reactions would under go a phenol/chloroform (pH 7-9) extraction in order to
denature all enzymes (Taq polymerase, Vent polymerase, and any RNases) present. A
phenol/chloroform extraction procedure will be explained further in the IRE purification
section. After the extraction the DNA fragments would be precipitated using 100%
isopropanol, then run over a NAP-25 Column (Pharmacia) that removes excess dNTPs
and salts, fractions would be checked by O.D. 260/280, and then the fractions would be
isopropanol precipitated and resuspend into water. The DNA was checked by O.D.
260/280 to determine the concentration, the DNA was now ready to be used in T7
transcription reactions, and it is stable for months at -20°C.
Results
The DNA obtained through PCR was checked for purity and correct size on
PAGE gels (see figures 28 and 29). For large scale amounts of DNA only primers BP37-RIR/BP-22-RIR and BP-25-RIR/BP-34-RIR were used for PCR. The DNA from the
large scale PCR reactions was phenol/chloroform extracted (see section in IRE
purification for procedure), isopropanol precipitated, and run over a desalting column
(NAP column, see section in IRE purification for procedure). It may be necessary to go
back and perform large scale PCR amplification of different HHR constructs in order to
obtain a homogenous IRE.
68
In vitro T l transcriptions
Enzymatic synthesis of RNA is commonly accomplished using the T l in vitro
transcription system developed by Uhlenbeck et al. (39). In an in vitro T l transcription,
double stranded DNA is converted into RNA by a T l RNA polymerase. The T l RNA
polymerase binds a specific sequence on the double stranded DNA called a T l promoter.
The polymerase begins synthesis immediately following the promoter and continues until
it reaches the end of the DNA strand. There are many components and conditions that
can affect the T7 reaction. The components and conditions in a T7 reaction usually
requires titrating individual components to identify optimal reaction conditions. Optimal
conditions should yield a high ratio of RNA to the amount of DNA template that is used.
A general T7 reaction is shown in Table 6, the current T7 reaction conditions are listed in
Table 7. Some general tips to consider when preparing for a T7 transcription reaction are
as follows:
1.
Prepare reactions at room temperature, since the presence of spermidine
above 4 mM (and MgCl2) can precipitate DNA at colder temperatures
((M0C) (47).
2.
T7RNP: Remove from the -20°C freezer immediately prior to use (only
component to be cold); Fresh DTT should be added to the T7RNP
stocks every 6 months.
3.
Normally, the reaction is allowed to proceed for I hour. However,
addition of another aliquot of T7RNP after this period followed by a
second hour of incubation will approximately double the yield (48).
69
Table 6: General T l transcription reaction conditions
Addition
order
3
COMPONENTS
OF THE REACTION
T7 "minimum" buffer, IOx
TrisHCl pH 8.1 @ 37C
(400 mM)
MgCl2 (40 M)
Spermidine (20 mM)
Triton X-100 (0.1 % v/v)
DTT (50 mM)
PEG 8k (27.5 % v/v)
2
Template DNA (nM)
4
Additional NaCl (2 M)
5
Additional KCl (1.5 M)
6
7
Additional PEG 8k (30 %)
Additional DMSO (100%)
8
PPiase (200 U/ml)
9
Additional BSA (1500 |ig/ml)
10
11
12
12
12
12
13
14
15
16
I
Comments
40mM final cone.; pH 8.4 @ 25C
4 mM in lx solution
2 mM in lx solution
0.01 % in lx solution
5 mM in lx solution
2.75 % in lx solution
5 - 500 nM final cone.
15 mM final cone.;
100 mM for short transcripts ( ~ 50 nt)
15 mM final cone.;
100 mM for short transcripts ( ~ 50 nt)
5 - 8 % final cone.
5 - 10 % final cone.
0.15-5 U/ml final cone.;
Roche 108 987
50 pg/ml final cone.; not for large scale
100 - 1000 U/ml final cone.;
Fischer Sci.
RNasin (40000 U/ml)
OR
SUPERase-In (Ambion)
5 - 5 0 mM final cone;
Additional DTT (1.5 M)
50 mM for short transcripts ( ~ 50 nt)
1 -5 mM final cone.
ATP (100 mM)
1 -5 mM final cone.
UTP (100 mM)
1 -5 mM final cone.
GTP (100 mM)
I - 5 mM final cone.
CTP (100 mM)
Additional Triton X-100 (1%) 0.01% final cone.
Final cone. = E [NTPs] +6
Additional MgCl2 (1.5 M)
Additional Spermidine (10 mM) 1 -4 mM final cone.
4000 - 80 000 U/ml final cone,
T l RNA pol (T7RNP)
optimum: 0.01 - 0.2 mg/ml
H2O
70
Table 7: Current T7 transcription reaction conditions for run-off T7 transcript using a
DNA template made from primers BP-25-RIR and BP-34-RIR
Addition
order
COMPONENTS
OF THE REACTION
T7 "minimum" buffer, IOx
TrisHCl pH 8.1 @ 37C
(400 mM)
MgCl2 (40 M)
Spermidine (20 mM)
Triton X-100 (0.1 % v/v)
DTT (50 mM)
_______ PEG 8k (27.5 % v/v)
Comments
3
40mM final cone.; pH 8.4 @ 25C
4 mM in lx solution
2 mM in lx solution
0.01 % in lx solution
5 mM in lx solution
2.75 % in lx solution
50 nM final cone, of run-off T7 template
0 mM final cone.
15 mM final cone.
8 % final cone.
7.5 % final cone.
0.15 U/ml final cone.;
Roche
2
4
5
6
7
Template DNA (nM)
Additional NaCl (2 M)
Additional KCl (1.5 M)
Additional PEG 8k (30 %)
Additional DMSO (100%)
8
PPiase (200 U/ml)
9
Additional BSA (1500 M-g/ml)
0 fig/ml final cone.; not for large scale
10
RNasin (40000 U/ml)
100 U/ml final cone. SUPERase-In (Ambion)
11
12
12
12
12
13
14
15
Additional DTT (1.5 M)
ATP (100 mM)
UTP (100 mM)
GTP (100 mM)
CTPGOO mM)
Additional Triton X-100 (1%)
Additional MgCl2 (1.5 M)
Additional Spermidine (10 mM)
16
T l RNA pol (T7RNP)
50 mM for short transcripts ( ~ 50 nt)
3 mM final cone.
3 mM final cone.
3 mM final cone.
3 mM final cone.
0.01% final cone.
18 mM final cone.
2 mM final cone.
2500 U/ml final cone.
T7 RNA pol Prep #5
I
H2O
71
4.
Large scale transcriptions often generate a considerable amount of a
white precipitate (presumably Mg2+ZPPi complexes) that can be reduced
or even avoided by the use of PPiase. The pyrophosphatase will also
prevent the backward reaction (49), i.e. RNA unpolymerization, thus
increasing yields. Otherwise, remove this precipitate either by
centrifugation or by addition of EDTA (RNase-free).
5.
The T7 transcription will generate a 5’ triphosphate extremity: Should
this extremity be removed with phosphatase?
6.
Although 40°C is the optimal temperature, incubation at 37°C generally
yields fewer incomplete transcripts without a negligible drop in yield
(46). The reaction works almost as well at 37°C as at 40°C, often with a
significant drop premature termination of transcription. In fact, reducing
the temperature to 30°C is often even more efficient at producing fulllength transcripts (>300 nt), but the yield of RNA drops to about 50%
(46).
7.
Under optimal conditions, the template and T7RNP concentrations are
limiting (45). Ensuring that none of the NTPs are not limiting, by
increasing the concentration of these NTPs can yield significantly more
full-length RNAs (long RNA, like mRNAs) (46).
8.
Concentrations that should be optimized are: template, T7RNP, NTPs,
MgCl2, NaCl, KCL, DTT, PEG, BSA, DMSO, and spermidine.
72
Results
T l reactions were run under the conditions listed in Table 6 and Table 7, and run
out on various denaturing PAGE gels. In figure 30 two different reactions were
performed and compared to the synthetic RNA. This figure indicates that the run-off IRE
transcript contains a major band at the same size as synthetic 30 nt IRE. The run-off IRE
contains incomplete transcription products (products less than 30 nt) and additional
transcription products (greater than 30 nt). While this transcription byproducts are
expected some of the additional transcription products were too large be from an
additional one or three nt (as expected with run-off transcriptions) added the IRE. The
HHR-IRE-HHR in the figure 30 is too large in size to effectively separate the bands on a
20% denaturing PAGE gel. The HHR-IRE-HHR transcript shows no significant band
that can be compared to the synthetic IRE, suggesting that the HHR-IRE-HHR can yet to
undergo significant cleavage. When a similar HHR-IRE-HHR transcription reaction is
run out on an 8% denaturing PAGE gel you can see multiple bands present. These
multiple bands could be from incomplete transcription products or some intial cleavage
of the HHRs. The reactions were also run out on a sequencing gel to determine exactly
what bands are present in a T7 reaction. Figure 32 shows that the major band from the
run-off reaction is IRE, while the HHR-IRE-HHR reactions show small amounts of IRE.
Incomplete and additional transcription products are present as well. These gels indicate
that the HHR cleavage needs to be optimized in order to get significant amounts of
homogenous IRE.
73
Additional
Uncharacterized
Transcription
Products
Incomplete
TranscriptiorT
Products
Fig. 30: Comparison of enzymatic synthesis versus chemical synthesis of IRE. T7 RNA
polymerase transcription reactions of run-off IRE (DNA from primers BP25/BP34), and
HHR-IRE-HHR (DNA from primers 1224/1233) run out on a 20% denaturing PAGE (7M
UREA) gel run at 300V.
5’ HHR-IRE-HHR 3’ 274 nt
3’ HHR: 161 nt
5’ HHR-IRE: 113 nt
5’ HHR: 83 nt
80I
60* *
IRE: 30 nt
Fig. 31: Various T7 RNA polymerase transcription reactions of HHR-IRE-HHR (made
from primers 1224/1233) run out on an 8% denaturing PAGE (7M UREA) gel run at
300V.
74
274 274
300nt %% nt,, nt 181
»
I
IOOnt w
I -
A B
C
"
Si
—
80nt
IOOnt
#*
60nt
%A
—
SOnt
**
e
40nt |j^
H
\
e
60nt
40nt
IRE
20nt
*
eXZt
IRE
T7
Transcription
rxns.
Run -off T7
Transcription
rxns.
Fig. 32: Two different T l transcription reactions that were labeled with Ci-S35UTP, single
nucleotide resolution can be obtained using a denaturing PAGE sequencing gel.
The gel on the left shows two different HHR-IRE-HHR, one is 274 nt (from primers
1224/1233) the other is 181 nt (from primers BP22/BP37).
The gel on the right shows run-off IRE 30 nt (from primers BP25/BP34). Lanes A:Runoff template/improved conditions for large scale rxn, B: Run-off template/initial
conditions for large scale rxn,C: Run-off template/conditions for small scale rxn
Reactions were run out on a 10% denaturing PAGE (7M UREA) sequencing gels run at
50 Watts for 20 cm.
75
IRE Purification
Following the synthesis of the IRE, subsequent purification is necessary to
remove any potential RNases that might degrade the IRE, to remove salts that may inhibit
IRP binding, and to separate the 30 nucleotide IRE from other RNA fragments.
Phenol/Chloroform Extractions
A Phenol/Chloroform extraction is used for initial purification of the RNA. Both
phenol and chloroform denature proteins and the denatured protein partitions in the lower
phenolic phase and/or at the interface, while nucleic acids remain in the aqueous phase.
Depending on the pH of the phenol, DNA will partition into either the organic phase or
the aqueous phase (50, 51). At pH > 7.0, both DNA and RNA partition into the aqueous
phase. On the contrary, at an acidic pH, DNA will be denatured and precipitate into the
organic phase and the interface, while the RNA remains in the aqueous phase.
Chloroform is mixed with phenol to reduce losses of nucleic acids at the interface.
The chloroform's ability to denature proteins increases the efficiency of the extraction.
Probably more, or at least as important, the phase separation is also enhanced by the
chloroform, hence allowing the removal of the aqueous phase with minimal
contamination from the organic phase and minimal loss of RNA.
Sometimes, for mRNA or rather long transcripts (> 300 nt), isoamyl alcohol
(IAA) is added to phenohchloroform (25:24:1) to reduce foaming, although there should
be no reason not to use it whatever the length of the transcript. After centrifugation, the
aqueous phase containing the nucleic acids is often re-extracted with an equal volume of
76
chlorofonrrisoamyl alcohol (50). This combination of extractions is thought to reduce the
loss of RNA due to the formation of insoluble proteimRNA complexes at the interface.
More important, oxidation products contained in the phenol or resulting from phenol
oxidation should be neutralized by adding 8-hydroxyquinoline (0.1 g/100 ml phenol) to
the phenol mixture prior to use.
Standard procedure for phenol extractions:
1. Determine the kind of phenol mixture to use (Table 8): e.g.
phenol:chloroform:IAA (25:24:1) pH 4.5 for mg amounts of RNA
2. First extraction: Mix I volume of the phenol mixture with I volume of the
solution containing the RNA, vortex for I min, put on ice for 10 min, centrifuge at
12,000 times g for 3 min, take the top (aqueous) phase.
3. Second extraction: Mix I volume of the aqueous phase with I volume of
chloroformTAA (24:1), vortex for I min, put on ice for 10 min, centrifuge at
12,000g for 3 min, take the top aqueous phase.
After the ERE has undergone the phenol/chloroform extraction, it is precipitated using an
equal volume of isopropanol (large scale) or 3 volumes ethanokO.l volumes of 3M
NaOAc pH 4.6 (small scale) and placed at -20°C overnight or at -80°C for one hour.
RNA is recovered by centrifugation at > 4000 g for one hour.
77
Table 8: Applications of pH specific phenol mixtures
* 8-hydroxyquinoline (0.1 g/100 ml phenol) should be added to the phenol mixtures prior
to use.
Saturated phenol
7.9 to
9.0
DNA/RNA
extraction
Saturated phenol: CHCh 7.9 to
9.0
(1:1)
DNA/RNA
extraction
Sat. phenol:CHCh:IAA
(25:24:1)
7.9
Saturated acid phenol
4.5
DNA/RNA
extraction
DNA/RNA
extraction
Sat. acid phenol:CHCh
(5:1)
or
RNA
extraction
4.5
Sat. acid
phenol:CHCh:IAA
(25:24:1)
Removal of
DNA
Often better for small scale (up to I
ml)
T7 transcriptions
(more NA in the aqueous phase)
Solution of choice for DNA extraction
Acidic pH moves DNA into the
organic phase
Acidic pH moves DNA (from in vitro
transcriptions) into the organic phase
CHC13 and IAA increase phenol
extraction efficiency
(reducing the loss, increasing the
phase separation and the protein
denaturation)
IAA prevents foaming when mixing
Nucleotide Removal
After the phenol/ chloroform extraction and precipitation, the free NTPS, and
salts need to be removed. We have tried a variety of buffer conditions, column types and
volumes, and flow rates. Gel filtration columns such as Superdex G25 (Amersham
Pharmacia), NAP columns (Amersham Pharmacia), and P-4 (Bio-Rad) are particularly
well suited for nucleotide removal. All of the gel filtration columns have pores that are
too small for the IRE, so the IRE elutes in the void volume of the column, while the
NTPs enter the matrix and are eluted after the void volume. The NAP column can handle
78
samples sizes of 0.5 ml to I ml, with a gel bed volume of 8.6 ml. A typical NAP column
protocol is as follows; wash column with 3x5 ml of HzO, apply 0.5 ml of sample to the
column, elute with 3x5 ml of HzO, and collect 0.5 ml fractions. The fractions are then
checked by O.D.. By looking at the spectral profile we are able to determine which
fractions contain IRE and which fractions contain NTPs (Fig. 33 and Fig. 34). The NAP
columns work better for larger RNAs than for shorter RNA, primarily due to the fact that
the column does not have large enough gel bed volume to effectively separate 30mer
RNA from NTPs (Fig. 33).
IRE run-off T7 reaction
N moles RNA
1.27E-07
NAP elution profile
160.0000
--------0 0 ( 2 6 0 nm) no dilution
140.0000 -
Theoritical profile
------- T7 transcript profile
120.0000
P C R product profile
100.0000
=B
80.0000
§
60.0000
N moles DNA
5.OOE-10
rNTPs2 profile
Yield
253
rN TPsl Profile
40.0000
20.0000
0.0000
10.0
-
11.0
12.0
13.0
20.0000
Elution volume (mL)
Fig. 33: Theoretical profiles of a IOml run-off T7 transcription reaction run over a NAP
column using O.D. 260 to indicate which peak contains the RNA and which peak
contains the NTPs. The theoretical profiling can be used to estimate the number of
moles of RNA produced.
79
10 mL T7 transcription NAP#1
(1 ml loaded; ~ 1/5 T7 reac)
0 0(26 0 nm) no dilution
—- Theoritical profile
— T7 transcript profile
rNTPI profile
— rNTP2 profile
NTPs
rNTP3 profile
— rNTP4 profile
— Phenol Profile
HHR-IRE-HHR
T7 Transcription
Reaction
El ution volume (mL)
N m o les RNA
1.25E-08
N m o le s DNA
1.00E-10
Yield
125
T h eo ritic al to tal
a m o u n t of RNA (^ g )
994.77
Fig. 34: Theoretical profiles of a IOml HHR-IRE-HHR T l transcription reaction run over
a NAP column using O.D. 260 to indicate which peak contains the RNA and which peak
contains the NTPs. The theoretical profiling can be used to estimate the number of moles
of RNA produced.
80
The Bio-Rad P4 column works the same way as the NAP column, but since it is not a
pre-packed column we are not limited by the gel bed volume. A P4 column with a bed
volume of 130 ml can handle sample volumes of up to 10 ml, and effectively separate
30mer IRE from NTPs (Fig. 35). The RNA elutes in the void volume while the NTPs
enter the matrix and thus elute later.
P ro file P4 B io G e 1 133 mL, 0 4 /0 8 /0 3
NTPs
OD260 no dil
o
30
Fig 35: P4 BioGel column loaded with IOml of 30mer ERE. First small peak is IRE,
while the second larger peak is NTPS.
Due to the polyanioinc character of nucleic acids it is possible to bind the DNA
and RNA to an anionic exchange resin. Therefore, it should be possible to find loading
conditions on the Mono Q column (Amersham Pharmacia), or CHT column {Bio-Rad)
such that the IRE will bind and the unincorporated NTPs would be washed away, hence
81
bypassing the gel filtration step. After initial experiments performed by Philippe Benas
(Montana State University) using an anion exchange column, MonoQ, further
developments were made to purify the IRE away from the HHRs and NTPs. The first
MonoQ run was done using HHR-IRE-HHR as the sample made from DNA using
primers BP-37/BP-22. The sample was loaded onto the column (0.5ml of a 10 ml
reaction). Buffer A (50mM Tris-HCl pH 6.8 (at 4°C)) was used to equilibrate the column
and the run starts at 100% buffer A, while buffer B: (50mM Tris-HCL pH 6.8 (at 4°C),.
IM NaCl) was used to start the gradient (0% to 100% buffer B). The absorbance reading
(1.0 Abs) was measured and fractions were collected at 0.5 ml intervals (Fig 36).
Analysis of column fractions using denaturing (7M UREA) PAGE gels indicated that
some NTPs copurified with the IRE, because the absorbance peak and the gel for the IRE
did not correlate. Thus, a second MonoQ column run was developed with a stepwise
elution holding the gradient constant at 10%, 20%, 30%, 40%, 50% buffer B for 5 ml
each. The result was a second peak of NTPs that eluted before the major NTP peak that
was indicated in the first MonoQ run (Fig. 37). Analysis of the column fractions using
denaturing (7M UREA) PAGE gels indicated that some NTPs were still associated with
the IRE, because the absorbance peak and the gel for the IRE did not correlate (figures
38, 39, and 40). The O.D. 260/280 ratio (1.2) and the spectral data indicated that the
excess signal strength for the IRE peak probably came from the NTPs and not another
buffer component.
82
Fig. 36: Mono Q run at 4°C of HHR-IRE-HHR (BP37/BP22). Loaded 0.5 ml of a 10 ml
T7 transcription reaction. Buffer A: 50mM Tris-HCl pH 6.8 (at 4°C) Buffer B: 50mM
Tris-HCL pH 6.8 (at 4°C), IM NaCl. Absorbance reading of 1.0
'.0
S
[RE
???
NTPS
/
10
J
I
HHRs
And
Full
Length
RNA
...
T
t
«
f-
t
*
u
M
f*
d
W
'I
"
/f
/I
V
u
Fig. 37: Mono Q run at 4°C of HHR-IRE-HHR (BP37/BP22) with step gradient. Loaded
0.5 ml of the same 10 ml T7 transcription reaction, and used same buffers as in Fig. 32.
Abs. 1.0
83
NTP Peak Fractions from Mono Q run
3
15 16 17
18
19
2 0 21
22
23
2 4 25
Fig. 38: 15% Denaturing (7M UREA) PAGE gel of fractions (3, 15-25) corresponding to
the peak containing NTPs from the MonoQ run in Fig. 33.
84
Fig. 40: 15% Denaturing (7M UREA) PAGE gel of fractions (35-43) corresponding to
the peak containing HHRs from the MonoQ run in Fig. 33.
Results
Removal of the NTPs from the transcription reactions was achieved by using NAP
columns and P-4 columns. The MonoQ column is able to bind the RNA and separate
most of the NTPs away from the RNA. It is evident from the gels that some NTPs co­
purify with the IRE peak seen in Figure 36 and 37 (the lanes on the gel containing the
IRE are very faint Fig. 39). This suggests that either the IRE is degraded rapidly or the
peak containing the IRE is receiving some extra abosorbance from co-purified NTPs.
The step gradient was able to remove NTPs (secondary peak of NTPs) suggesting that the
conditions can be optimized further to remove all the NTPs.
85
IRE Quality Control
During and after IRE purification it is necessary to check the quality of the IRE.
There are a variety of ways to accomplish this goal, but it usually involves UV
absorbance or denaturing PAGE gels. UV absorbance is often selected because it offers a
quick and accurate method for determining RNA concentration and protein
contamination. Small denaturing PAGE gels are intermediate in time consumption and
allow one to estimate the relative amounts of different signal transcripts. At times, single
nucleotide resolution is necessary in which case sequencing gels loaded with radiolabeled
RNA are used. Sequencing gels require several days to complete, but may be worth the
time and effort in order to resolve the products at the single nucleotide level.
UV-absorbance
By checking the UV absorbance at 260 nm and 280 nm the quantity and quality of
IRE can be determined. Absorbance at 260 nm translates into 40/tg/ml per I GD, while
the GD 260/280 ratio can tell you if there are any contaminants in the sample, typically
1.9 is a good ratio. Also, by following the absorbance at 260 nm during a run on the
sizing columns or anion exchange columns, it can tell you where the RNA and NTPs
elute. UV absorbance is a quick and effective method for determining the quantity of
IRE present in a sample. It is usefull to determine an accurate concentration of IRE
before trying to from the complex with the IRPs.
86
Mini PAGE
A quick way to check for IRE presence and quality is to use a mini denaturing PAGE gel.
The gel usually contains 7M urea, IX TBE, 20% acrylamide and is run for one hour at
250-300 volts. This gel doesn’t require any radiolabeled sample to be read, ethidium
staining followed by UV illumination works fine. Radiolabeled samples can be run on
the mini gels, if the acrylamide concentration is too high the gels will crack during the
drying process. If the acrylamide concentration is too low the resolution will be poor.
These gels typically work best for non-radio-labeled samples that need to be checked for
the presence of IRE.
Sequencing Gels
To determine RNA quality and quantity one of the first things to try is to run
radiolabeled RNA in a denaturing PAGE sequencing gel. Typically the ERE is either
spiked during a T7 transcription reaction with radioactive OC-35S-UTP (use standard T7
reaction procedure but add the desired amount of radiolabel, typically IOp Cl/ pi for a
50pl reaction volume) or is kinased with Y-33P-ATP. The radiolabeled RNA is then
loaded with 95% formamide gel loading dye onto a 7M urea, 20% formamide, IX TBE,
10 % acrylamide sequencing gel and run for 20-25 cm at 50 watts. The gels were dried
for 2 hrs (I 1/2 hours of 80°C heat and 2 hours under vacuum), and allowed to expose in
a Kodak phosphoimager plate (Bill Dyer, Montana State University) overnight (at least 4
hours). The plate is then placed in a Molecular Dynamics phosphoimager (Al Jesaitis,
Montana State University), and analyzed using ImageQuant software. A sequencing gel
87
is used in order to obtain single nucleotide resolution. From the gel, the quality of the
IRE can be determined by comparing the different bands, and the quantity can be
determined (during a T7 transcription labeling) by measuring the density of the bands.
The disadvantages of running a sequencing gel is that it is time consuming taking
anywhere from 2-3 days to complete, the gels are too thin to ethidium stain and check by
UV, and radiolabeling may cause gel artifacts.
The sequencing gels also allow one to follow the HHR cleavage reaction, and to
assay for optimal cleavage conditions (Fig. 41) (52, 53). Two 50/ri a-[35S]-UTP
radiolabeled T7 transcription reactions were made for HHR cutting experiments, Cutting
experiments were conducted on a 274 nt HHR-IRE-HHR (DNA template from primers
1224/1233), and 181 nt HHR-IRE-HHR (DNA template from primers 1224/BP22). The
T7 reactions mixtures were passed through a G-25 microcentrifuge spin column
(Amersham) to remove any salts and excess NTPs. One third of each reaction had
varying concentrations of MgCla added. The reactions contained OmM, 30mM, and 60
mM MgCla respectively, and were heated at 70°C for 30 min (52, 53). The reactions were
then run out on a 10% denaturing (7 M urea) sequencing gel at 50 watts for 20 cm to
separate the IRE. The sequencing gel showed the disappearance of the full length 274 nt
and 181 nt long RNA, and the appearance of a 30 nt long IRE. The sequencing gel also
showed that the 5’HHR ribozyme had cut and was visible at 83 nt in every lane, but no
3’HHR could be seen. The 274 nt long RNA should have generated a 161 nt long 3’HHR
and the 18Imer should have generated a 68 nt long 3’HHR once the cut was complete,
but neither could be seen with certainty on the gel (Fig. 41).
88
274
nt
«
an
■
I
I
100a*
274
nt
274 nt
0 mM
181
nt
_
60
181 n t
0 mM
60
30
30
Mt
m
80 * .
**
6 0 jhj|
F
M
Esi
\
/
5’ HHR
40*
m
Fig. 41: Cutting experiment on HHR-IRE-HHR 274 mer (primers 1224/1233),
and HHR-IRE-HHR ISlmer (primers 1224/BP22). OmM, 30mM, and 60 mM
MgCl2 were each added to a 20^1 T7 reaction that was heated at 70°C for 30 min.
and run out on a 10% denaturing (7 M urea) sequencing gel at 50 watts for 20 cm.
Since, the reactions were heated in the presence of MgCl2, the RNA could have
undergone chemical cleavage, which would cause the lack of expedited bands. Once the
reactions volumes were increased to greater than 1ml, the cutting experiments never
repeated the success that was seen in the small scale reactions. This could be due to a
89
number of reasons; including reaction volume, RNA concentration, MgCla concentration,
temperature effects, etc.
Band Shift Assays
One way to check for IRE/IRP binding activity is to run a band shift assay. In a
band shift assay the IRE is radio labeled with either a-S35-UTP or y-P33-ATP.
Following Elizabeth Leibold’s (University of Utah) band shift protocol, radio labeled
RNA and IRP (I or 2) are bound together at O0C for 30 minutes in a solution containing:
IOmM Hepes pH 7.6, 3mM MgCl2, 40mM KCL, 5% glycerol, and ImM dithiothreitol
(DTT) (24). The IRE/IRP complex is run on a pre-electrophoresed 4% non-denaturing
polyacrylamide gel (19:1 acryl/bis acryl ratio), 0.25X TBE and run at 4°C for 60 min at
100 volts. The gel is dried, exposed to a phosphoimager plate, and developed. The band
shift assay does not determine how much IRP or IRE is active; it only shows that the IRP
can bind IRE (Fig 42). From the literature on IRP bandshift assays IRE/IRP-1 is always
shifted higher than IRE/IRP-2. The bandshift assay did show that our “homemade” run­
off T7 IRE could bind to IRP-I and some of the IRP-2 (disappearance of the free IRE
band and faint appearance of the IRE/IRP-2 band) indicating that the proteins and the
IRE were active. The synthetic IRE was able to shift with IRP-I and IRP-2, once again
the IRE/IRP-2 shift was not as well defined as the IRE/IRP-2. IRP-2 not binding the
IREs in this bandshift assay is probably due to the IRP-2 protein being a week old, we
have seen the IRP-2 bind IRE stronger in previous bandshift assays. The band shift assay
is a useful tool in determining if you have any active IRE and IRP.
90
;
I
2
3
4
5
6
Fig. 42: Bandshift Assay (4% native PAGE gel, 0.25xTBE, run at 100 V for I hr.) of
radiolabeled synthetic IRE verses radiolabeled T7 run-off IRE being bound by IRP-I
and IRP-2. Lane I: run-off IRE, Lane 2: run-off IRE/IRP-1, Lane 3: run-off IRE/IRP2, Lane 4: commercial IRE, LaneS: commercial IRE/IRP-1, Lane 6: commercial
IRE/IRP-2.
Conclusions
Sequencing gels and UV absorbance are good ways to check for RNA quality and
quantity, while mini PAGE gels are quick and easy to use they just don’t provide the
resolution necessary to analyze small RNA fragments such as the IRE. Band shift assays
indicated that IRP-1, and IRP-2 could bind homemade IRE to some extent.
91
CHAPTER 4
Conclusions
The IRE/IRP story is a challenging one that hopefully this research will help to
unravel. Since, the Hentze and Kuhn 1996 review, few major advances in the story have
been made. Hopefully when the structure is of the IRP/IRE is solved the some of the
major questions will be answered. There were a few small advances made by this
research.
TRP-2 minus degradation domain
While the expression of IRP-2 minus the degradation domain, has not yielded any
substantial results the idea is sound. If the IRP-2 mutants express as well or better than
the wild-type IRP-2 in yeast, the mutant IRPs should be easier to purify and crystallize.
Currently IRP-2-ADD is expressing at 0.5mg/L in yeast cell culture. The method for
obtaining the mutants is a powerful technique that could be used to generate more
mutants of IRP-2 that might express and behave better.
IRE
The current protocol for T7 transcription reactions is using the run-off T7
transcript, and it is yielding approximately 1-2 mg of IRE per 60 ml of reaction. This
yield can still be improved upon, but for right now it is generating enough IRE to be able
to move forward. The HHR IRE constructs are still coming along, initial cleavage results
looked promising at the small scale T7 transcription reaction stage, but when the T7
92
transcription reaction volumes were increased the HHR on longer cleaved under the same
conditions as the small scale T l transcription reactions. The yield of the T7 reaction is
supposed to increase with the increased size of the transcript, but if HHR cleaves away
the IRE only 50% of the time, usable IRE is approximately the same as the run-off
transcript. The only advantage of the HHR IRE method is that the DRE is more suitable
for crystallization, by itself and for making DRE/DPR complex.
The area of greater concern is purifying away the NTPs after the reaction is
completed. The column that removes the nucleotides the best is the P-4 Column by BioRad. Although, it still appears that some NTPs are still co-purifying with the ERE,
because the absorbance readings and the gels don’t agree with each other. The
absorbance indicates that there should be 2-3 times the amount to IRE that is indicated by
denaturing PAGE. There might be a couple of reasons for this discrepancy besides NTPs
co-purifying with the ERE: one is that the IRE may be to small to be effectively stained
with ethidium bromide so there should be more present than can be seen; another is that
the absorbance reading is off due to instrument or user error; and the IRE may be
degraded by RNases or some sort of chemical cleavage. It may turn out that some NTPs
will always migrate with the IRE, leading to an overestimate of IRE. Still the majority of
NTPs are separated from the ERE through size exclusion chromatography.
The good news is that the ERE that is being made is at the correct size
(approximately 32 bases for run-off transcript), and can actively bind ERE, as is evident
by the sequencing gels and band shift assays. Most recently, Philippe Benas performed
some complex formation/ purification experiments for IRE/ERP-1 that resulted in IRE
binding IRP in a ratio of 1:1. It would be nice if the mini-PAGE gels would be able to
93
resolve the IRE better, because it would be nice to have a quick way to look for the IRE
verses using a sequencing gel. The main problems in getting the gel to have good
resolution are IRE size, gel size, and salt concentration in the IRE sample.
Future Experiments
While the research performed in the lab has been a step forward, there are many
experiments left to try. Something that should be explored further is the HHR-IRE
cleavage experiments, currently the lab is obtaining enough usable IRE by run-off
transcription that the cleavage experiments are not a necessity. Another area of focus is
to continue to work on IRP-2-ADD expression in yeast. Further purification of the IRE is
necessary as well; while all of the columns currently used separate most of the NTPs and
salts away from the IRE it would be nice to see distinct separate peaks for both of them.
The main area of focus that has to be explored further is to form the IRE/IRP-l complex
and the IRE/IRP-2 complex so that crystallization trials can be completed. Once crystals
of the complex are obtained, structure determination can begin.
94
REFERENCES CITED
1. Hentze, M. W., & Kuhn, L. C. (1996) Proc. Natl. Acad. ScL U. S. A. 93, 81758182
2. Hentze, M. W., Caughman, S. W., Rouault, T. A., Barriocanal, J. G., Dancis, A.,
Harford, J. B. & Klausner, R. D. (1987) Science 238,1570-1572.
3. Brown, P. H., Daniels-McQueen, S., Walden, W. E., Patino, M. M., Gaffield, L.,
Bielser, D. & Thach, R. E. (1989) J. Biol. Chem. 264,13383-13386.
4. Zahringer, J., Baliga, B. S. & Munro, H. N. (1976) Proc. Natl. Acad. ScL USA 73,
857-861.
5. Brown, P. H., Daniels-McQueen, S., Walden, W- E., Patino, M. M., Gaffield, L.,
Bielser, D. & Thach, R. E. (1989) J. Biol. Chem. 264, 13383-13386.
6. Goossen, B. & Hentze, M. W. (1992) Mol. Cell. Biol. 12, 1959-1966.
7. Gray, N. K. & Hentze, M. W. (1994) EMBO J. 13, 3882-3891.
8. Casey, J. L., Hentze, M. W., Koeller, D. M., Caughman, S. W., Rouault, T. A.,
Klausner, R. D. & Harford, I. B. (1988) Science 240, 924-928.
9. Casey, J. L., Di Jeso, B., Rao, K., Klausner, R. D. & Harford, J. B. (1988) Proc.
Natl. Acad. ScL USA 85, 1787-1791.
10. Koeller, D. M., Casey, J. L., Hentze, M. W., Gerhardt, E. M., Chan, L.-N.,
Klausner, R. D. & Harford, J. B. (1989) Proc. Natl. Acad. ScL USA 86, 35743578.
11. Ward, J. H., Kushner, I. P. & Kaplan, J. (1982) J. Biol. Chem. 257,10317-10323.
12. Rao, K. K., Shapiro, D., Mattia, E., Bridges, K.& Klausner, R. (1985) Mol. Cell.
Biol. 5, 595-600.
13. Rao, K. K., Harford, J. B., Rouault, T., McClelland, A., Ruddle, F. H. &
Klausner, R. D. (1986) Mol. Cell. Biol. 6, 236-240.
14. Milliner, E. W., Neupert, B. & Kiihn, L. C. (1989) Cell 58, 373-382.
15. Casey, I. L,, Koeller, D. M., Ramin, V. C., Klausner, R. D. & Harford, J. B.
(1989) EMBO J. 8, 3693-3699.
95
REFERENCES CITED - CONTINUNED
16. Seiser, C., Posch, M., Thompson, N. & Ku'hn, L. C. (1995) J. Biol. Chem. 270,
29400-29406.
17. May, B. K., Bhasker, C. R., Bawden, M. J. & Cox, T. C. (1990) Mol. Biol. Med.
7, 405-421.
18. Dunkov, B. C., Zhang, D., Choumarov, K., Winzerling, I. J. & Law, J. (1995)
Arch. Insect Biochem. Physiol. 29, 293-307.
19. Hentze, M. W., Caughman, S. W., Casey, J. L., Koeller, D. M., Rouault, T. A.,
Harford, I. B. & Klausner, R. D. (1988) Gene 72, 201-208.
20. McCallum, S. A., & Pardi, A. (2003) J. Mol. Biol. 326,1037-105
21. Barton, H. A., Eisenstein, R. S., Bomford, A. & Munro, H. N. (1990) J. Biol.
Chem. 265, 7000-7008.
22. Beaumont, C., Leneuve, P., Devaux, I., Scoazec, J.-Y., Berthier, M., Loiseau, M.N., Grandchamp, B. & Bonneau, D. (1995) Nat. Genet. 11,444-446.
23. Henderson, B. R., Menotti, E., Bonnard, C. & Kuhn, L. C. (1994) J. Chem. 269,
17481-17489.
24. Leibold, E. A. & Munro, H. N. (1988) Proc. Natl. Acad. Set USA 85, 2171-2175.
25. Henderson, B. R., Seiser, C. & Kuhn, L. C. (1993) J. Biol. Chem. 268, 2732727334.
26. Hentze, M. W., Rouault, T. A., Harford, J. B. & Klausner, R. D. (1989) Science
244, 357-359.
27. Kennedy, M. C., Mende-Mueller, L., Blondin, G. A. & Beinert, H. (1992) Proc.
Natl. Acad. Sci. USA 89,11730-11734.
28. Hirling, H., Henderson, B. R. & Kuhn, L. C. (1994) EMBO J. 13, 453-461.
29. Neupert, B., Menotti, E. & Ku'hn, L. C. (1995) NucleicAcids Res. 2579-2583.
30. Rouault, T. A. (2001) Pediatr. Neurol. 25(2), 130-137
31. Iwai K., Klausner R.D., & Rouault TA. (1995) EMBO J. 14(21), 5350-5357
96
REFERENCES CUED - CONTINUNED
32. Guo, B., Yu, Y. & Leibold, E. A. (1994) J. Biol Chem. 269, 24252-24260.
33. Kang D.K., Jeong J., Drake S.K., Wehr N.B., Rouault TA., & Levine R.L. (2003)
J. Biol. Chem. 278(17), 14857-14864
34. Goessling L.S., Mascotti D.P., & Thach R.E. (1998) J. Biol. Chem. 273(20),
12555-12557
35. Grabill C., Silva A.C., Smith S.S., Koretsky A.P., & Rouault TA. (2003) Brain
Res. 971(1), 95-106.
36. Ho S.N., Hunt H.D., Horton R.M., Pullen J.K., & Pease L.R. (1989) Gene 77(1),
51-59
37. Horton R.M., Hunt H.D., Ho S.N., Pullen J.K., & Pease L.R. (1989) Gene 77(1),
61-68
38. Ke Y., Wu J , Leibold E.A., Walden W.E., & Theil E.C. (1998) JBzoZ Chem.
273(37), 23637-23640
39. Milligan J.F., Groebe D.R., Witherell G.W., & Uhlenbeck O.C. (1987) Nucleic
Acids Res. 15(21), 8783-8798
40. McConnell T.S., Herschlag D., & Cech T.R. (1997) Biochemistry 36(27), 82938303
41. Doudna JA, & Cech TR. (2002), Nature 418(6894), 222-228
42. Ambion (2002) Tech. Notes 9(3)
http://www.ambion.com/techlib/tn/93/9316.html
43. Condon, C. (2003), Microbiol Mol Biol Rev. 67(2): 157-174.
44. Chamberlin, M., & Ring, J. (1973), J. Biol. Chem. 248, 820-825
45. Arnold S., Siemann M., Schamweber K., Werner M., Baumann S., & Reuss M.
(2001), Biotechnol. Bioeng. 72, 548-561
46. Yisraeli, J.K., & Melton, DA. (1989), Methods Enzymol. 180, 42-50
47. Hoopes, B.C., & McLure, W.R. (1981), NucleicAcids Res. 9, 5493
97
REFERENCES CITED - CONTINUNED
48. Cunningham, P.R., & Ofengand I. (1990), Biotechniques 9, 713-714
49. Melton D.A., Krieg PA., Rebagliati M.R., Maniatis T., Zinn K., & Green M.R.
(1984), Nucleic Acids Res.12, 7035-7056
50. Wallace, D.M. (1987), Methods Enzymol. 152, 33-40
51. Wallace, D.M. (1987), Methods Enzymol. 152, 41-48
52. Herschlag D, & Khosla M. (1994,) Biochemistry 33(17), 5291-5297
53. Leibold E.A., & Munro H.N. (1988) Proc Natl Acad Sci U. S. A. 85(7), 21712175.
MONTANA STATE UNIVERSITY - BOZEMAN
*
Download